ARTHRITIS & RHEUMATISM
Vol. 62, No. 2, February 2010, pp 359–368
© 2010, American College of Rheumatology
Methotrexate Polyglutamate Concentrations Are
Not Associated With Disease Control in
Rheumatoid Arthritis Patients Receiving
Long-Term Methotrexate Therapy
Lisa K. Stamp,1John L. O’Donnell,2Peter T. Chapman,2Mei Zhang,2Jill James,2
Christopher Frampton,3and Murray L. Barclay1
Objective. There are limited data suggesting that
methotrexate polyglutamate (MTXGlu) concentrations
can guide MTX dosing in patients with rheumatoid
arthritis (RA). The aim of this study was to define a
therapeutic range of red blood cell (RBC) MTXGlun
concentrations (where n refers to the number of gluta-
mate groups), including threshold values for efficacy
and adverse effects in patients receiving long-term oral
Methods. A cross-sectional study of 192 patients
receiving oral MTX was undertaken. Disease activity
was assessed by the swollen and tender joint counts, the
C-reactive protein level, and the Disease Activity Score
in 28 joints (DAS28). High disease activity was defined
as a DAS28 of >3.2. A standardized questionnaire
regarding common MTX adverse effects was completed.
Results. The MTX dosage was significantly higher
in patients in whom the swollen joint count and DAS28
were higher. The MTXGlu4, MTXGlu5, MTXGlu3–5, and
MTXGlu1–5concentrations were significantly higher in
patients with high disease activity. After correction for
age, the estimated glomerular filtration rate, and the
MTX dosage, the association remained significant for
MTXGlu5. RBC folate concentrations were significantly
higher in the group with high disease activity. There was
no association between any MTXGlunconcentration
and adverse effects.
Conclusion. In contrast to other studies, the
results of the present study did not show a relationship
between the MTXGlunconcentration and reduced dis-
ease activity in patients with RA who were receiving
long-term MTX therapy. However, disease activity was
influenced by the RBC folate level, which may be a more
important factor than MTXGlunconcentrations for
disease control. In accordance with the findings of
previous studies, we were unable to show a relationship
between MTXGlunconcentrations and adverse effects.
Prospective studies will be important to determine
whether there is a role for measuring MTXGlunconcen-
trations in patients receiving long-term treatment with
Methotrexate (MTX) is the gold standard treat-
ment against which newer disease-modifying antirheu-
matic drugs (DMARDs) are compared, because of its
established track record in rheumatoid arthritis (RA).
The dosage of MTX required by individual patients
varies greatly and is unpredictable. The decision by a
clinician to increase the dosage is dependent on the
assessment of disease activity, the accepted upper limit
of drug dosing, and adverse effects.
In patients with RA, both a trend toward a
dose-response relationship (1,2) and no association be-
tween dose and response have been reported for MTX
(3). Serum MTX concentrations have not correlated
with disease activity (4). Recent studies suggest a nega-
tive correlation between long-chain red blood cell
(RBC) MTX polyglutamate (MTXGlun) concentrations
ClinicalTrials.gov identifier: ACTRN012606000275561.
Supported by Health Research Council of New Zealand and
Arthritis New Zealand.
1Lisa K. Stamp, FRACP, PhD, Murray L. Barclay, FRACP,
MD: University of Otago, and Christchurch Hospital, Christchurch,
New Zealand;2John L. O’Donnell, FRACP, FRCPA, Peter T. Chap-
man, FRACP, MD, Mei Zhang, PhD, Jill James, BN: Christchurch
Hospital, Christchurch, New Zealand;3Christopher Frampton, PhD:
University of Otago, Christchurch, New Zealand.
Address correspondence and reprint requests to Lisa K.
Stamp, FRACP, PhD, Department of Medicine, University of Otago,
Christchurch, PO Box 4345, Christchurch 8140, New Zealand. E-mail:
Submitted for publication March 29, 2009; accepted in revised
form October 9, 2009.
(where n refers to the number of glutamate groups) and
disease activity (3,5,6). Studies have also examined
whether polymorphisms of genes encoding the enzymes
involved in MTX transport and/or the folate pathway
may help predict the response and/or adverse effects
associated with MTX. However, to date, there is no clear
consensus regarding the most important polymorphisms
(for review, see ref. 7).
Despite its widespread use, the exact mechanism
of action of MTX remains unclear. Following adminis-
tration and absorption, serum MTX concentrations fall
rapidly (8), and MTX is transported into a variety of
cells including RBCs, white blood cells, hepatocytes, and
synoviocytes via the reduced folate carrier (RFC). Intra-
cellularly, glutamate moieties are added by folylpolyglu-
tamate synthetase (FPGS), and MTX is retained as
MTXGlun. Terminal MTX glutamates are removed by
?-glutamyl hydrolase, returning MTX to its monogluta-
mate form, which is rapidly transported out of the cell by
multidrug-resistant proteins. The intracellular concen-
trations of MTXGlunare therefore related to the bal-
ance of activity between these 2 enzymes. Ingested
folates are also transported into cells by the RFC and
folate receptor and compete with MTX as a substrate for
polyglutamation by FPGS. Thus, high concentrations of
intracellular folates result in a decrease in MTX poly-
MTX has several potential adverse effects. The
more common, but less serious, adverse effects that
often limit a patient’s willingness to continue MTX
treatment include nausea and lethargy. The more seri-
ous adverse effects, which can limit the dosage or
require MTX discontinuation, include hepatotoxicity
and myelotoxicity. Because of the inability to predict the
occurrence of these adverse effects, patients are re-
quired to have regular blood tests. Two small studies in
patients with RA demonstrated no clear association be-
tween MTXGlunconcentrations and adverse effects (5,11).
The aim of this study was to determine whether a
therapeutic range of MTXGlunconcentrations could be
defined in order to guide adjustment of oral MTX
dosing, including threshold values for efficacy and ad-
PATIENTS AND METHODS
Ethics approval was obtained from the Upper South B
Regional Ethics Committee, New Zealand. Written informed
consent was obtained from each patient.
Patients and study protocol. This cross-sectional study
was undertaken in a single center in Christchurch, New
Zealand. Patients ?18 years age with RA, as defined by the
American College of Rheumatology (formerly, the American
Rheumatism Association) (12), were recruited. Patients were
required to have been receiving oral MTX for ?3 months, and
the dosage must have been stable for ?1 month prior to study
entry. A preference was given to patients receiving MTX
monotherapy. Concomitant therapy with nonsteroidal antiin-
flammatory drugs (NSAIDs) and prednisone was allowed. A
change in the MTX dosage, introduction of another DMARD,
a change in the oral prednisone dosage, and receipt of an
intraarticular steroid injection within 1 month prior to enroll-
ment were not allowed.
Clinical assessment. Patient demographic and clinical
details were collected on standardized data collection forms.
Disease activity was assessed using the swollen joint count
(SJC), the tender joint count (TJC), the modified Health
Assessment Questionnaire (M-HAQ) (13), the physician’s
global assessment of disease activity, and the patient’s assess-
ment of pain and global assessment of disease activity. To
avoid interobserver variability, the SJC and TJC were deter-
mined by a single trained observer. The physician’s assessment
of response to MTX was measured using a 100-point visual
analog scale (VAS), where 0 ? excellent response and 100 ?
poor response. The Disease Activity Score in 28 joints
(DAS28) (14) was calculated, and low disease activity was
defined as a DAS28 of ?3.2 (15). The Clinical Disease Activity
Index (CDAI) (16) and the Simplified Disease Activity Index
(SDAI) (17) were also calculated.
A standardized questionnaire related to common
MTX adverse effects in the month preceding study entry was
developed. Adverse effects were grouped into 3 categories, as
follows: 1) gastrointestinal (GI; nausea, vomiting, diarrhea,
mouth ulcers, and decreased appetite), 2) central nervous
system (CNS; fatigue, loss of concentration, headache, dizzi-
ness, blurred vision, sleep disturbance, and weepiness), and 3)
other (hair loss, cough, fever, and shortness of breath).
Laboratory measurements. Standardized laboratory
assessments included a complete blood cell count, liver func-
tion tests, and determination of the creatine level, the eryth-
rocyte sedimentation rate (ESR), the C-reactive protein
(CRP) level, and the RBC folate level. Estimated creatinine
clearance (the estimated glomerular filtration rate [eGFR])
was calculated using the Modification of Diet in Renal Disease
study equation (18).
MTXGlu terminology and measurement. MTX, the
parent drug, contains 1 glutamate moiety and is referred to as
MTXGlu1.MTXGlu1and the products of intracellular gluta-
mation (MTXGlu2, MTXGlu3, MTXGlu4, and MTXGlu5) are
collectively referred to as MTX polyglutamates (MTXGlun).
The terms MTXGlu3–5and MTXGlu1–5refer to the sum of
each measured RBC MTXGlunconcentration.
Trough MTXGlunconcentrations were measured by
high-performance liquid chromatography, as previously de-
scribed (19). Results were normalized to an RBC count of 8 ?
1012cells, so that results were comparable and not confounded
by changes in RBC counts between individuals. All samples
were analyzed in duplicate, and the mean concentration of
each RBC MTXGlunfrom each sample was used.
Statistical analysis. The relationship between the
MTX dosage and each disease activity measure was assessed
360STAMP ET AL
using Spearman’s correlation coefficient. The relationships
between MTXGlunconcentrations and disease activity mea-
sures were first quantified using correlation coefficients. Sub-
sequently, the relationships were further explored using covari-
ate analyses that included variables known to affect disease
activity (e.g., age, sex, duration of RA, and prednisone use) as
covariates. Differences between the groups with low disease
activity and high disease activity in terms of MTXGluncon-
centrations, demographic and clinical features, and treatment
measures were tested using independent t-tests, nonparametric
Mann-Whitney U tests, chi-square tests, and Fisher’s exact
tests, as appropriate.
Further covariate analyses exploring the independent
association of MTXGlunconcentrations with disease activity
were undertaken using age, eGFR, MTX dosage, and RBC
folate level as covariates. The relationships between the pres-
ence of adverse effects and MTX dosage and MTXGlun
concentrations were tested using independent t-tests. Covari-
ate analyses including other factors that might contribute to
adverse effects (use of other DMARDS or NSAIDs, duration
of MTX treatment, and RBC folate levels) were undertaken to
establish the independent association between adverse effects
and MTX dosage and MTXGlunconcentrations. The associa-
tion between MTXGlunconcentrations and other laboratory
measurements was tested using Pearson’s correlation coeffi-
cient. Two-tailed P values less than 0.05 were considered
Demographics. Two hundred patients were re-
cruited between October 2005 and February 2008. An
ESR value was not obtained in 8 patients, and these
patients were excluded, leaving 192 in the final analysis.
Of the 192 patients, 72.9% were women, and the mean
age was 60.5 years (range 18–84 years). The mean
duration of RA was 10.5 years (range 0.25–53 years).
Among the 192 patients, 25.5% had rheumatoid nod-
ules, 63% had radiographic erosions, 80.7% were rheu-
matoid factor positive, and 76.4% were anti–cyclic cit-
rullinated peptide (anti-CCP) antibody positive. The
median dosage of MTX was 15 mg/week (range 5–25),
and all but 1 patient received folic acid at a dosage of 5
mg/week, given 3–4 days after the dose of MTX. Patients
had been receiving MTX for a median of 3 years (range
0.25–19 years) prior to study entry and had been receiv-
ing MTX at the study-entry dosage for a median of 12
months (range 1–240 months). Eighty-six of the 192
patients (44.8%) were receiving NSAIDs, and 20 pa-
tients (10.4%) were receiving another DMARD (4 sul-
fasalazine [SSZ], 9 hydroxychloroquine [HCQ], 5 both
SSZ and HCQ, 1 leflunomide [LEF], and 1 both LEF
and HCQ). Fifty-nine patients (30.7%) were receiving
oral prednisone at a mean dosage of 5.2 mg/day (range
Relationship between MTX dosage, MTXGlun
concentrations, and disease activity. There was a signif-
icant association between the MTX dosage and RBC
MTXGlu3, MTXGlu4, MTXGlu5, MTXGlu1–5, and
MTXGlu3–5concentrations (P ? 0.0001 for all). How-
ever, there was no association between the MTX dosage
and the MTXGlu1or MTXGlu2concentration (20).
The MTX dosage was higher in those patients
with a higher SJC (r ? 0.19, P ? 0.006), physician’s
global assessment score (r ? 0.3, P ? 0.0001), DAS28
(r ? 0.15, P ? 0.03), physician-rated response to MTX
(r ? 0.27, P ? 0.0001), patient’s global assessment score
(r ? 0.19, P ? 0.007), CDAI (r ? 0.25, P ? 0.0005), and
SDAI (r ? 0.25, P ? 0.0005). There was no association
between the MTX dosage and the TJC, M-HAQ score,
CRP level, ESR, pain on a VAS, or fatigue on a VAS.
Thus, despite increasing dosages of MTX, disease re-
mained active in some patients.
RBC MTXGlunprofile. MTXGlu1, MTXGlu2,
and MTXGlu3were detectable in all patients, while
MTXGlu4was detectable in 178 (92.7%) of 192 patients,
and MTXGlu5was detectable in 148 (77.1%) of 192
patients. MTXGlu3was the predominant polyglutamate,
accounting for a mean of 35.8% of the total MTXGlun.
MTXGlu1accounted for 21.7%, MTXGlu2accounted
for 20.5%, MTXGlu4accounted for 14.5%, and MTX-
Glu5accounted for 7.5%.
Association between the RBC MTXGlunconcen-
tration and disease activity measures. Univariate analy-
sis revealed a significant association between MTXGlu4,
MTXGlu5, MTXGlu1–5, and MTXGlu3–5concentrations
and the SJC, the physician’s global assessment, the
physician’s assessment of response to MTX, the DAS28,
the CDAI, and the SDAI, such that higher disease
activity was associated with higher MTXGlunconcentra-
tions (Table 1). There was also a positive correlation
between the MTXGlu3concentration and the physi-
cian’s global assessment score, the DAS28, and the
SDAI (Table 1). There was a significant association
between MTXGlu1–5concentrations and the M-HAQ
score. There was no association between any other
MTXGlunconcentration and the M-HAQ, the CRP
level, or the patient’s assessment of pain and global
assessment of disease activity on a VAS (Table 1).
Several variables are known to affect disease
activity and outcome measures in RA, including age, sex,
duration of RA, smoking, anti-CCP antibody status,
concomitant treatment with corticosteroids, NSAIDs, or
other DMARDs, and folate status. After adjustment
for these variables, patients with a higher TJC, a higher
SJC, and a higher DAS28 still had significantly higher
MTX POLYGLUTAMATES IN RA 361
MTXGlu5concentrations (P ? 0.05 for all). Similarly,
patients with greater fatigue and higher M-HAQ scores
had significantly higher MTXGlu2concentrations (P ?
0.05). Taken together, these data suggest that physicians
may increase the MTX dosage in the setting of ongoing
disease activity in an effort to improve the drug effect, as
would be expected. Patients receiving a higher dosage of
MTX and having higher disease activity may therefore
be nonresponders to MTX.
RBC MTXGlunconcentrations in the group with
high disease activity and the group with low disease
activity. One hundred fourteen patients (59%) were
defined as having low disease activity (DAS28 ?3.2),
and 78 (41%) were categorized as having high disease
activity. Other than longer disease duration in the group
with high disease activity, there were no significant
differences in demographic and treatment variables be-
tween these 2 groups (Table 2).
The MTXGlu4, MTXGlu5, MTXGlu3–5, and
MTXGlu1–5concentrations were higher in the group
with high disease activity compared with the group with
low disease activity (Figure 1). In this cohort, we previ-
ously showed that age, the eGFR, and the MTX dosage
account for most variation in MTXGlunconcentrations
(20). After correction for these variables, there was no
significant difference between the group with high dis-
Univariate analysis of RBC MTXGlunconcentrations and disease activity variables*
VAS Fatigue CDAISDAI
* Values are Spearman’s correlation coefficients. The visual analog scale (VAS) was based on a 0–100 scale. RBC ? red blood cell; MTXGlu ?
methotrexate polyglutamate; M-HAQ ? modified Health Assessment Questionnaire; CRP ? C-reactive protein; DAS28 ? Disease Activity Score
in 28 joints; VAS ? visual analog scale; CDAI ? Clinical Disease Activity Index; SDAI ? Simplified Disease Activity Index.
† P ? 0.01.
‡ P ? 0.05.
Characteristics of the patients with RA according to the level of disease activity*
Low disease activity
(n ? 114)
High disease activity
(n ? 78)P
Age, mean (range) years
No. of men/no. of women
eGFR, mean ? SEM ml/minute/1.72 m2
Rheumatoid factor positive
Anti-CCP antibody positive
Duration of RA, mean (range) years
Duration of RA ?20 years
Any other DMARD
Dosage, mean (range) mg/day
MTX dosage, median (IQR) mg/week
No. of months receiving dosage, median (IQR)
76.9 ? 1.4
75.7 ? 1.9
* Except where indicated otherwise, values are the number (%). Low disease activity was defined as a
28-joint Disease Activity Score (DAS28) of ?3.2; high disease activity was defined as a DAS28 of ?3.2.
RA ? rheumatoid arthritis; eGFR ? estimated glomerular filtration rate; anti-CCP ? anti–cyclic
citrullinated peptide; NSAIDs ? nonsteroidal antiinflammatory drugs; DMARD ? disease-modifying
antirheumatic drug; MTX ? methotrexate; IQR ? interquartile range.
† Only 77 of the patients were assessed.
‡ Only 113 of the patients were assessed.
362 STAMP ET AL
ease activity and the group with low disease activity in
the concentration of MTXGlu1(P ? 0.76), MTXGlu2
(P ? 0.47), MTXGlu3(P ? 0.38), MTXGlu4(P ? 0.08),
MTXGlu1–5(P ? 0.11), or MTXGlu3–5(P ? 0.08).
However, MTXGlu5concentrations remained signifi-
cantly higher in the group with high disease activity (P ?
0.02). MTXGlu5was undetectable in 25% of the group
with low disease activity compared with 19% of the
group with high disease activity (P ? 0.31).
When patients were grouped according to broad
ranges of each MTXGlunconcentration, similar percent-
ages of patients were classified as having low disease
activity, irrespective of the MTXGlunconcentration
It has been suggested that patients in whom the
MTXGlu3–5concentration is ?60 nmoles/liter are more
likely to have a good response to MTX (3). Among our
patients with a DAS28 of ?3.2, 43% had an MTXGlu3–5
concentration of ?60 nmoles/8 ? 1012RBCs, and 57%
had an MTXGlu3–5concentration of ?60 nmoles/8 ?
Figure 1. Methotrexate polyglutamate (MTXGlu) concentrations in patients with low disease activity (Disease Activity Score in 28 joints [DAS28]
?3.2) and patients with high disease activity (DAS28 ?3.2). Each dot represents an individual patient. Bars show the mean. RBC ? red blood cell.
Figure 2. Percentage of patients with low disease activity (Disease Activity Score in 28 joints [DAS28] ?3.2) according to ranges of each
methotrexate polyglutamate (MTXGlu) concentration. RBC ? red blood cell.
MTX POLYGLUTAMATES IN RA363
1012RBCs. Among patients with a DAS28 of ?3.2, 31%
had an MTXGlu3–5concentration of ?60 nmoles/8 ?
1012RBCs, and 69% had an MTXGlu3–5concentration
of ?60 nmoles/8 ? 1012RBCs (Table 3).
RBC folate concentrations and disease activity.
The mean ? SEM RBC folate concentration was higher
in the group with high disease activity compared with
that in the group with low disease activity (786.9 ? 31.2
nmoles/liter versus 664.2 ? 27.4 nmoles/liter; P ? 0.002).
Current smokers had lower mean ? SEM RBC folate
concentrations compared with current nonsmokers
(611.9 ? 44.7 nmoles/liter versus 743.6 ? 24.5 nmoles/
liter; P ? 0.008).
After correction for RBC folate level in addition
to age, eGFR, and MTX dosage, the group with high
disease activity still had higher MTXGlu5concentrations
(P ? 0.046). However, the remaining MTXGlunconcen-
trations were not significantly different between the
group with high disease activity and the group with low
disease activity (data not shown).
Association between MTX dosage, RBC MTX-
Glunconcentration, and MTX adverse effects. One
hundred forty-one patients (73.4%) reported ?1 ad-
verse effect, 118 (61.4%) reported ?1 CNS-related
adverse effect, 81 (42.2%) reported ?1 GI-related ad-
verse effect, and 53 (27.6%) reported ?1 other adverse
effect. There was no association between the MTX
dosage and the presence of any adverse effect (P ?
0.53), a CNS-related adverse effect (P ? 0.31), a GI-
related adverse effect (P ? 0.70), or other adverse effect
(P ? 0.39).
There was no association between RBC MTX-
Glunconcentrations and the presence of any adverse
effect, GI-related adverse effect, or CNS-related adverse
effect. There was a trend toward association between
“other” adverse effects and higher RBC MTXGlu3(P ?
0.07), MTXGlu4(P ? 0.052), and MTXGlu3–5(P ?
Allowing for factors that may contribute to ad-
verse effects, such as concomitant treatment with other
DMARDs or NSAIDs and the duration of time receiv-
ing MTX and folate, there was still a trend toward an
association between the presence of “other” adverse
effects and higher MTXGlu3(P ? 0.08) and MTXGlu4
(P ? 0.08) concentrations. There was no association
between any other MTXGlunconcentrations and the
presence of any adverse effect, GI-related adverse ef-
fect, or CNS-related adverse effect (data not shown).
Many patients experience adverse effects early in
the course of therapy, necessitating discontinuation of
MTX. We therefore examined the relationship between
the presence of adverse effects and MTXGlunconcen-
trations in 35 patients receiving MTX for ?12 months.
There were no associations between any MTXGlun
concentration and adverse effects.
There was no association between the RBC fo-
late concentration and the presence of any adverse
effects (P ? 0.27), CNS-related adverse effects (P ?
0.17), GI-related adverse effects (P ? 0.29), or other
adverse effects (P ? 0.34).
Relationship between MTXGlunconcentrations
and laboratory variables. There was a significant nega-
tive correlation between hemoglobin and MTXGlu1–5
(P ? 0.03), MTXGlu2(P ? 0.002), and MTXGlu3(P ?
0.04) concentrations. There was a positive correlation
between the mean cell volume and the MTXGlu1–5(P ?
0.03) and MTXGlu1(P ? 0.003) concentrations and a
trend toward a correlation with MTXGlu3(P ? 0.07).
There was no association between any MTXGluncon-
centration and the platelet, neutrophil, or lymphocyte
count (data not shown). There was no association be-
tween any MTXGlunconcentration and the aspartate
Classification of the patients according to the DAS28 and RBC MTXGlu3–5concentrations*
No. (%) of
DAS28 ?3.2 (n ? 114)
DAS28 ?3.2 (n ? 78)
* Low and high concentrations of methotrexate polyglutamate (MTXGlu) were defined as ?60 nmoles/
8 ? 1012red blood cells (RBCs) and ?60 nmoles/8 ? 1012RBCs, respectively. DAS28 ? Disease Activity
Score in 28 joints.
364 STAMP ET AL
aminotransferase, alanine aminotransferase, or serum
albumin concentration (data not shown).
MTX remains the “anchor” drug in the manage-
ment of RA. Although it is evident that early intensive
DMARD therapy can prevent joint damage, a key
challenge that remains is determining whether an indi-
vidual patient will respond to a particular drug, and at
what dosage. In practice, patients are begun on a low
dosage of MTX (7.5–10 mg/week), with upward titration
according to response until the maximum dosage is
reached (20–25 mg/week) (21). Such an approach may
be a waste of time in nonresponsive or MTX-resistant
patients. The ability to perform a blood test to deter-
mine whether further dosage escalation will provide
therapeutic benefit would be a major advance. Such an
advance would aid decisions regarding dosage increases
for MTX monotherapy versus changing to alternative or
combination DMARD or biologic therapies. It has been
suggested that RBC MTXGlunconcentrations can fulfill
such a role and help guide MTX dosing in RA.
The lack of an association between MTXGlun
concentrations and reduced disease activity in this cross-
sectional study in patients receiving long-term MTX
treatment was surprising and contrasts with earlier re-
ports. Somewhat unexpectedly, we observed a positive
correlation between MTXGlu5and the SJC, TJC,
M-HAQ score, and DAS28, which persisted despite
correction for age, sex, duration of RA, smoking, anti-
CCP antibody status, concomitant treatment with ste-
roids, NSAIDs or other DMARDS, and RBC folate
The clinical significance of these results is uncer-
tain, but they suggest that there is a group of patients
in whom disease remains active despite higher MTX
dosages and higher RBC MTXGlu5concentrations. It
would seem logical that the MTX dosage had been
increased in order to increase the drug effect in these
patients with persistently active disease that failed to
respond to MTX at lower dosages. This is supported by
the findings that MTX dosages and MTXGlunconcen-
trations were higher in patients with more active disease,
perhaps defining an MTX-resistant population. This
reflects a confounding-by-indication scenario that is a
consequence of MTX as the first-line DMARD in our
practice, unless there is a specific contraindication. The
results, therefore, reflect our local practice of increasing
the MTX dosage to the maximum prior to the introduc-
tion of alternative DMARDs. The presence of patients
with high disease activity and high MTXGlunconcentra-
tions is likely to mask a positive association between
MTXGlunconcentrations and disease control in those
patients in whom disease does not respond to MTX.
We acknowledge that the cross-sectional study
design has limitations. It dictated that many patients had
been receiving MTX, often at stable dosages for long
durations. The population studied was heterogeneous,
and management was under the direction of 3 different
rheumatologists who made therapeutic decisions based
on perceptions of disease activity and patient choices
rather than a predefined treatment protocol. Such a
study design does not enable us to address the question
of whether increasing the MTX dosage, thus increasing
MTXGlunconcentrations, results in additional disease
control. A prospective study would enable examination
of both the response to MTX and the occurrence of the
more severe adverse effects that necessitate discontinu-
ation of therapy. Such a study would enable comparisons
between different dosing strategies and responses to
MTX therapy but would be lengthy and logistically
difficult. However, clinicians are frequently faced with
patients presenting with active disease who have already
been receiving treatment for a period of time. The key
clinical question in this “real life” setting is whether to
increase the dosage of MTX or add or change to another
DMARD. In this regard, a cross-sectional study design
examining the relationship between MTXGlunconcen-
trations and disease activity, such as the design used in
this study, provides useful information.
It is possible that the duration of MTX treatment
prior to measurement of the MTXGlunconcentration
may be an important factor. In a study of 48 patients
with RA who were commencing MTX therapy, respond-
ers had higher MTXGlunconcentrations at 6 months
compared with nonresponders (40 ? 18 nmoles/liter
versus 27 ? 6 nmoles/liter), although significance could
not be determined because of the low number of non-
responders (11). Patients with RBC MTXGlunconcen-
trations of ?20 nmoles/liter at 3 months were more
likely to have had a smaller change in the DAS28.
Another study of 40 patients with RA who were com-
mencing MTX therapy showed that “erythrocyte (Ery)-
MTX” concentrations were higher in responders com-
pared with nonresponders (P ? 0.013) (22). Thus, in
early RA, it has been shown that higher MTXGlun
concentrations are associated with lower disease activity,
but this relationship may wane as disease becomes
established due to alterations in the biologic process of
RA over time (2).
In patients receiving long-term MTX therapy, the
MTX POLYGLUTAMATES IN RA365
role of measuring the MTXGlunconcentration seems
less clear. In a study of 65 patients receiving MTX for
?2 months, erythrocyte MTX concentrations were
higher in responders compared with nonresponders
(60.92 ? 19.1 versus 21.47 ? 10.67; P ? 0.0001). In that
study, the total duration of therapy with MTX was not
reported, and response was based on the physician’s
global assessment including history, joint examination,
and a comparison with the patient’s condition at an
earlier time point (5). In a cross-sectional study of 108
patients with RA who were receiving MTX for ?3
months (median 65 months, range 3–266 months),
higher RBC MTXGlu3concentrations were associated
with lower TJCs (P ? 0.04), lower SJCs (P ? 0.021), and
a lower score for the physician’s global assessment (P ?
0.001) but not with the ESR or the M-HAQ score. In
addition, MTXGlu concentrations of ?60 nmoles/liter
were associated with a 14-fold higher likelihood of a
VAS score ?2 cm for the physician’s assessment of the
response to MTX (3). In a study of 226 patients with RA
who were receiving MTX for ?3 months (median 51
months, interquartile range 19–97 months), patients
with MTXGlu concentrations of ?60 nmoles/liter were
4-fold more likely to have a poor response (as defined by
the physician’s assessment of the patient’s response to
The cross-sectional nature of this study means
that it is not possible to determine whether those
patients with a DAS28 of ?3.2 and an MTXGlu3–5
concentration of ?60 nmoles/liter may have achieved a
similar level of disease control at lower MTXGlun
concentrations. Although the MTX dosage is generally
increased until a response is achieved, it is possible that
some patients may have responded to lower dosages.
Dosage escalation typically occurs at 2–6-week intervals,
and our previous work has shown that it takes much
longer than this for steady-state concentrations to be
achieved (19); therefore, the dosage may have been
increased unnecessarily in some patients.
In all of the cross-sectional studies to date, in-
cluding the present study, the minimum duration of
MTX therapy has been 3 months, with the dosage being
stable for ?1 month. Given that it can take up to 6
months to reach steady state after commencing MTX
therapy and after dosage adjustments, it is likely that
steady-state concentrations may not have been achieved
in some patients. However, using the same cohort of
patients, we have shown that a longer duration of
therapy with MTX was associated with higher RBC
MTXGlu4and MTXGlu5concentrations only in a uni-
variate analysis, and this association did not persist in
the multivariate model including age, dosage, and renal
Our data suggest that there is a group of patients
(31%) with a high DAS28 and low MTXGlunconcen-
trations in whom a dosage increase could be considered.
If such an approach were undertaken, we would antici-
pate that ?50% of patients may experience an addi-
tional reduction in disease activity.
In our cohort, 41% of patients had high disease
activity. It was our preference to enroll patients receiv-
ing MTX monotherapy, in order to minimize confound-
ing from the use of other DMARDs. Although there is
no disagreement that low disease activity and remission
are the desired outcomes, such outcomes are not
achieved in all patients. A variety of factors contribute to
ongoing disease activity, including patient treatment
choices and availability of alternative DMARDs. In
addition, some patients may have been in the process of
receiving dosage escalations or, alternatively, may have
presented with a flare of arthritis after previously having
had good disease control.
We have shown that patients with high disease
activity have significantly higher RBC folate concentra-
tions than those with low disease activity. This is consis-
tent with results of a previous study in which reduced
RBC folate levels were associated with improvement in
disease control in patients with RA who were beginning
therapy with MTX (11). Our data suggest that RBC
folate levels may be a more important factor than
MTXGlunconcentrations for disease control in patients
receiving MTX. This possibility warrants further inves-
In patients with a poor response to therapy,
reduced drug compliance should be considered. Some
laboratory results, such as an increase in the mean red
cell volume (MCV), may give an indication of compli-
ance, in that the MCV should rise gradually after
starting treatment with MTX (23). It has been suggested
that MTXGlu3–5concentrations can also be used to help
determine compliance. However, we have previously
shown that many weeks or months are required for
MTXGlu3–5to reach steady state and to be eliminated
(19). Thus, only patients who are consistently noncom-
pliant over many weeks may exhibit low concentrations.
Consistent with previous reports (3,5), we have
shown no association between MTXGlunconcentrations
and MTX-related adverse effects. Because there is no
validated questionnaire for the adverse effects of MTX,
we developed a questionnaire, using expert opinion. The
questionnaire addressed the presence or absence of
well-recognized but often underappreciated and under-
366STAMP ET AL
reported adverse effects of MTX. We acknowledge that
many of these adverse effects are nonspecific, and that
placebos may be associated with such adverse effects.
The majority of the patients in this study had
received MTX for several years. Therefore, the cross-
sectional nature of the study would have excluded many
patients who could not tolerate MTX and discontinued
treatment early, thereby making it more difficult to
detect a relationship between MTXGlunconcentrations
and adverse effects and factors that predict adverse
effects. A prospective study is needed to determine the
relationship between MTXGlunconcentrations and se-
rious adverse effects such as myelotoxicity, pneumonitis,
and/or hepatotoxicity, which usually necessitate MTX
discontinuation or dosage reduction. In addition, it is
not clear whether RBC MTXGlunconcentrations reflect
accumulation in other tissues such as liver or bone
marrow. In this regard, Ahern et al examined liver
biopsy specimens obtained from 18 patients with RA
who were receiving MTX and observed increased con-
centrations of MTX, MTXGlu2, and 2,4 diamino-N10-
methylpteroic acid (another MTX metabolite) in 3 pa-
tients with hepatic fibrosis (24).
Using the same questionnaire that was used in
this study, a small study of 18 patients with inflammatory
bowel disease demonstrated higher MTXGlu5concen-
trations in patients experiencing ?1 MTX-related ad-
verse effect, and MTXGlu4and MTXGlu5concentra-
tions were higher in those experiencing GI adverse
effects (25). It is possible, due to the small number of
patients, that this was a Type I error. Alternatively, there
may be differences in the perception or pattern of
adverse effects between different patient groups.
In summary, the results of our cross-sectional
study do not support earlier findings of a correlation
between MTXGlunconcentrations and the control of
RA disease activity. This result was unexpected but may
reflect differences in established versus early disease,
selection of patients tolerant to MTX, dosage increases
in MTX-resistant patients, RBC folate status, differ-
ences in intracellular drug metabolism, or other as-yet-
The best way to clarify this situation would be to
conduct a long-term prospective study, serially examin-
ing the relationships between MTX dosage, MTXGlun
concentrations, RBC folate status, and effects/adverse
effects after commencing treatment. Such a long-term
study may help explain the apparent differences in
findings between studies, the optimal time for measuring
MTXGlunconcentrations after commencing MTX ther-
apy, and the relationship between the MTXGluncon-
centration and loss of disease control in patients receiv-
ing MTX. Furthermore, such a study may help to
determine whether there is clinical benefit in increasing
the dosage of MTX in patients who are not responding
to treatment. Careful consideration will need to be given
to the study design, given the long time required for
MTXGlunconcentrations to reach a steady state and the
need for rapid disease control.
Although we have shown no relationship between
MTXGlunconcentrations and adverse effects, a pro-
spective study would clarify whether there is a relation-
ship between MTXGlunconcentrations and serious
MTX-related adverse effects that necessitate MTX dis-
continuation. Finally, RBC folate status has been shown
to be an important influence on MTX efficacy. Further
studies should help to address the optimal folate regi-
men and RBC folate levels with concomitant MTX
We gratefully acknowledge the assistance of Jan Ipen-
burg, Rheumatology Clinical Nurse Specialist, in assisting with
patient data collection.
All authors were involved in drafting the article or revising it
critically for important intellectual content, and all authors approved
the final version to be published. Dr. Stamp had full access to all of the
data in the study and takes responsibility for the integrity of the data
and the accuracy of the data analysis.
Study conception and design. Stamp, O’Donnell, Chapman, Framp-
Acquisition of data. Stamp, O’Donnell, Chapman, Zhang, James.
Analysis and interpretation of data. Stamp, O’Donnell, Chapman,
James, Frampton, Barclay.
1. Furst DE, Koehnke R, Burmeister LF, Kohler J, Cargill I.
Increasing methotrexate effect with increasing dose in the treat-
ment of resistant rheumatoid arthritis. J Rheumatol 1989;16:
2. Seideman P. Methotrexate: the relationship between dose and
clinical effect. Br J Rheumatol 1993;32:751–3.
3. Dervieux T, Furst D, Lein DO, Capps R, Smith K, Walsh M, et al.
Polyglutamation of methotrexate with common polymorphisms in
reduced folate carrier, aminoimidazole carboxamide ribonucleo-
tide transformylase, and thymidylate synthase are associated with
methotrexate effects in rheumatoid arthritis. Arthritis Rheum
4. Lafforgue P, Monjanel-Mouterde S, Durand A, Catalin J, Acqua-
viva PC. Lack of correlation between pharmacokinetics and effi-
cacy of low dose methotrexate in patients with rheumatoid arthri-
tis. J Rheumatol 1995;22:844–9.
5. Angelis-Stoforidis P, Vajda FJ, Christophidis N. Methotrexate
polyglutamate levels in circulating erythrocytes and polymorphs
MTX POLYGLUTAMATES IN RA367
correlate with clinical efficacy in rheumatoid arthritis. Clin Exp
6. Dervieux T, Furst D, Lein DO, Capps R, Smith K, Caldwell J, et
al. Pharmacogenetic and metabolite measurements are associated
with clinical status in rheumatoid arthritis patients treated with
methotrexate: results of a multicentred cross sectional observa-
tional study. Ann Rheum Dis 2005;64:1180–5.
7. Ranganathan P. An update on methotrexate pharmacogenetics in
rheumatoid arthritis [review]. Pharmacogenomics 2008;9:439–51.
8. Tishler M, Caspi D, Graff E, Segal R, Peretz H, Yaron M. Synovial
and serum levels of methotrexate during methotrexate therapy of
rheumatoid arthritis. Br J Rheumatol 1989;28:422–3.
9. Jolivet J, Faucher F, Pinard MF. Influence of intracellular folates
on methotrexate metabolism and cytotoxicity. Biochem Pharmacol
10. Kennedy DG, van den Berg HW, Clarke R, Murphy RF. The
effect of leucovorin on the synthesis of methotrexate poly-gamma-
glutamates in the MCF-7 human breast cancer cell line. Biochem
11. Dervieux T, Greenstein N, Kremer J. Pharmacogenomic and
metabolic biomarkers in the folate pathway and their association
with methotrexate effects during dosage escalation in rheumatoid
arthritis. Arthritis Rheum 2006;54:3095–103.
12. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF,
Cooper NS, et al. The American Rheumatism Association 1987
revised criteria for the classification of rheumatoid arthritis.
Arthritis Rheum 1988;31:315–24.
13. Pincus T, Summey JA, Soraci SA Jr, Wallston KA, Hummon NP.
Assessment of patient satisfaction in activities of daily living using
a modified Stanford Health Assessment Questionnaire. Arthritis
14. Prevoo ML, van ‘t Hof MA, Kuper HH, van Leeuwen MA, van de
Putte LB, van Riel PL. Modified disease activity scores that
include twenty-eight–joint counts: development and validation in a
prospective longitudinal study of patients with rheumatoid arthri-
tis. Arthritis Rheum 1995;38:44–8.
15. Van Gestel AM, Haagsma CJ, van Riel PL. Validation of rheu-
matoid arthritis improvement criteria that include simplified joint
counts. Arthritis Rheum 1998;41:1845–50.
16. Aletaha D, Nell VP, Stamm T, Uffmann M, Pflugbeil S, Machold
K, et al. Acute phase reactants add little to composite disease
activity indices for rheumatoid arthritis: validation of a clinical
activity score. Arthritis Res Ther 2005;7:R796–806.
17. Smolen JS, Breedveld FC, Schiff MH, Kalden JR, Emery P, Eberl
P, et al. A simplified disease activity index for rheumatoid arthritis
for use in clinical practice. Rheumatology (Oxford) 2003;42:
18. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D, et al.
A more accurate method to estimate glomerular filtration rate
from serum creatinine: a new prediction equation. Ann Intern
19. Dalrymple JM, Stamp LK, O’Donnell JL, Chapman PT, Zhang M,
Barclay ML. Pharmacokinetics of oral methotrexate in patients
with rheumatoid arthritis. Arthritis Rheum 2008;58:3299–308.
20. Stamp LK, O’Donnell JL, Chapman PT, Zhang M, Frampton C,
James J, et al. Determinants of red blood cell methotrexate
polyglutamate concentrations in rheumatoid arthritis patients re-
ceiving long-term methotrexate treatment. Arthritis Rheum 2009;
21. Visser K, Katchamart W, Loza E, Martinez-Lopez JA, Salliot C,
Trudeau J, et al. Multinational evidence-based recommendations
for the use of methotrexate in rheumatic disorders with a focus on
rheumatoid arthritis: integrating systematic literature research and
expert opinion of a broad international panel of rheumatologists in
the 3E initiative. Ann Rheum Dis 2009;68:1086–93.
22. Hornung N, Ellingsen T, Attermann J, Stengaard-Pedersen K,
Poulsen JH. Patients with rheumatoid arthritis treated with meth-
otrexate (MTX): concentrations of steady-state erythrocyte MTX
correlate to plasma concentrations and clinical efficacy. J Rheu-
23. White DH, Chapman PT, O’Donnell JL, James J, Frampton C,
Stamp LK. Lack of association between elevated mean red cell
volume and haematological toxicity in patients receiving long-term
methotrexate for rheumatoid arthritis. Intern Med J 2009. E-pub
ahead of print.
24. Ahern MJ, Kevat S, Hill W, Hayball PJ, Harley H, Hall PM.
Hepatic methotrexate content and progression of hepatic fibrosis:
preliminary findings. Ann Rheum Dis 1991;50:477–80.
25. Brooks AJ, Begg EJ, Zhang M, Frampton CM, Barclay ML. Red
blood cell methotrexate polyglutamate concentrations in inflam-
matory bowel disease. Ther Drug Monit 2007;29:619–25.
368 STAMP ET AL