вход по аккаунту


The safety and efficacy of the use of methotrexate in long-term therapy for rheumatoid arthritis.

код для вставкиСкачать
Twenty-nine patients participated in a prospective study of the safety and efficacy of oral methotrexate
in the treatment of refractory rheumatoid arthritis.
Pallients received a mean dosage of 12.4 mg weekly over
a mean duration of 29.1 months. All patients had liver
biopsies at baseline, 2 years, and annually thereafter.
Patients improved significantly by all clinical measures
of eflicacy after 1 month; maximum improvement
tendetd to occur after approximately 6 months of therapy. Radiographs showed improvement of erosive disease in 7 of 11 patients measured. There was a significant reduction in mean prednisone dosage. Four
patienits required an increase in the dosage of methotrexate after prolonged therapy, because of declining
clinical response. Toxicity was noted at some time in 26
of 29 patients (90%), but reactions universally became
mild and tolerable after adjustment of the dosage. No
significant hepatotoxicity was found in 60 sequential
liver biopsies, although elevated transaminase levels
were noted at some time in 20 of 29 patients (70%).
Although the use of folate antagonists for the
treatment of rheumatoid arthritis (RA) was first suggestedl in 1951 ( l ) , only recently have blinded clinical
trials established the short-term efficacy of methotrexate (MITX) in the treatment of patients with active RA
From the Division of Rheumatology, Albany Medical College and the Department of Radiology, Albany Medical Center
Hospital, Albany, New York.
Supported in part by NIH grant M01-RR-00749-13 and a
grant from Lederle Laboratories.
Joel M. Kremer, MD; Joong Kil Lee, MD.
Address reprint requests to Joel M. Kremer, MD, Division
of Rheumatology, Albany Medical College, Albany, NY 12208.
Submitted for publication October 21, 1985; accepted in
revised form January 2, 1986.
Arthritis and Rheumatism, Vol. 29, No. 7 (July 1986)
that has failed to respond to standard therapeutic
approaches (2,3). Because there has been extensive
experience with the use of MTX in the treatment of
psoriasis (4), most practitioners who used MTX to
treat patients with arthritis adopted the dermatologic
guidelines which had identified individuals at high risk
for the development of hepatotoxicity. Thus, patients
with a history of alcoholism, liver or renal disease, or
diabetes mellitus were excluded from trials of the use
of MTX in RA (5-8).
Although clear recommendations for monitoring potential hepatotoxicity with liver biopsies
emerged from the experience with methotrexate in
psoriasis patients (9), most investigators following RA
patients who were undergoing long-term therapy with
this drug have not routinely performed biopsies (5-7).
The problem of monitoring MTX hepatotoxicity is
compounded by the well-established lack of correlation between liver function test results and hepatic
changes noted in patients with psoriasis (4,9). Furthermore, other investigators have reported hepatic abnormalities in patients with rheumatoid arthritis who have
not received therapy (lCL12). Clearly, it is difficult to
interpret hepatic cellular changes in an RA patient
receiving MTX, who has not had a baseline biopsy for
Several authors have commented on the rapid
therapeutic benefit seen with use of MTX (2,7,13), but
it is not clear if these initial benefits are long-term, or
the disease becomes refractory to the drug after an
initial favorable therapeutic response, as has been
suggested previously (14). Other questions which have
not been adequately studied include whether there is
an effect on disease progression (as seen on radiographs), the frequency and management of long-term
side effects, and the relationship of toxicity to duration
of therapy. A genetic predisposition to both a favorable ,therapeuticresponse to MTX (2) and to hepatotoxicity (15) has been suggested but has not been
explored in long-term trials.
We have addressed some of these issues in an
open, long-term prospective study with baseline and
followup liver biopsies.
Patients. Patients seen at Albany Medical College
from November 1980 through September 1984 who had
either failed to respond to treatment with hydroxychloroquine, gold, or penicillamine, or who had had a toxic
reaction to one of these drugs which prevented continuation
of therapy, were asked to participate. Twenty-nine RA
patients who fulfilled the American Rheumatism Association
criteria for definite or classic disease (16) entered the study.
All patients had received gold salt therapy, 24 patients (83%)
had received hydroxychloroquine, 24 patients (83%) had
received penicillamine, 2 patients (7%) had received
azathdoprine, and 1 patient (3%) had received cyclophosphamide. One patient had received an 11-week course of oral
methotrexate which was discontinued 5 months prior to
entry into the study because of mild nausea and anxiety
about continuing the drug. The previous response to MTX in
this individual was judged “fair” by the rheumatologist who
had prescribed the drug. Patients had discontinued all disease-modifying drugs for at least 1 month prior to initiating
MTX therapy.
Other requirements for entry were the presence of
active disease defined by at least 3 of the following criteria:
2 3 swollen joints, 2 6 tenderjoints, 2 4 5 minutes of morning
stiffniess, and a Westergren erythrocyte sedimentation rate
(ESF!) 2 2 8 mmhour.
Excluded from the study were patients judged to be
in functional class IV by the criteria of Steinbrocker et a1 (17)
or those patients with a history of cancer, liver or renal
disease, white blood cell counts of <3 ,S00/mm3, or platelet
counts <1.5 x ioS/mrn3.
Twenty-seven patients were white, 1 was black, and
1 was of partly American Indian ancestry. Each patient was
instructed that he or she was allowed no more than 3 oz of
alcohol per week or 16 oz of beer per week. Patients were
questioned on a monthly basis about their compliance with
this regulation.
Study design. All patients signed informed consent
prior to entry in the study and agreed to the proposed
schedule of liver biopsies. All patients had baseline liver
biopsies before beginning MTX therapy. Biopsies were
repeated after 2 years, which corresponded to a 1.5-gm
cumulative dosage of MTX in most patients, and annually
thereafter. Liver biopsies were read by a pathologist who
had no knowledge of the duration of MTX therapy or of prior
liver biopsy results. Biopsy specimens were processed with
homatoxylin and eosin, trichrome, and reticulum staining.
All biopsies were waded according to the classification
previously devised by Roenigk et a1<9) to evaluate hepatic
changes in psoriatic patients receiving MTX: class I (normal)
= mild fatty infiltration, mild nuclear variability, mild portal
inflammation; class I1 = moderate to severe fatty infiltration, moderate to severe nuclear variability, and moderate to
severe portal tract expansion, portal tract inflammation, and
necrosis; class I11 = mild fibrosis, tract inflammation, and
necrosis; class I11 = mild fibrosis, formation of fibrotic septa
extending into lobules; class IV = cirrhosis.
Patients were begun on a regimen of 15 mg of MTX
taken orally in 1 dose each week. Elderly patients (>65 years
old) and 2 patients weighing <50 kg were begun on a regimen
of 7.5 mg orally once each week. All patients were seen
monthly by a physician for evaluation of toxicity and were
asked to rate adverse effects they experienced as either mild,
moderate, or severe. If patients had nausea, abdominal
discomfort, loose stools, oral ulcerations, or mouth soreness
which they rated as moderate or severe, the same dosage of
MTX would be scheduled at 3 consecutive 12-hour intervals
once a week. If the adverse effect persisted, the dose of
MTX was lowered by 2.5 mg each week until these reactions
either disappeared or became mild and tolerable.
Clinical assessment. Clinical disease variables were
determined by the same physician at baseline, 1 month, 3
months, and every 3 months thereafter, and consisted of the
number of swollen joints; the number of joints with tenderness on pressure, or pain on passive motion (or both); mean
grip strength for both hands; 15-meter (SO-foot) walking time
without assistive devices; the duration of morning stiffness;
the physician’s assessment of pain and disease activity
graded as 0 (asymptomatic), 1 (mild), 2 (moderate), 3 (severe), or 4 (very severe); and the patient’s assessment of
pain and disease activity graded on the same scale.
Laboratory assessment. The following laboratory test
values were obtained at baseline and every 2 weeks during
the first year of MTX therapy and then monthly, if there
were no abnormalities during this period of time: complete
blood count, including platelets, serum glutamic oxaloacetic
transaminase (SGOT), gamma glutamyl transferase (GGT),
alkaline phosphatase, albumin, total bilirubin, total protein,
and creatinine. Urinalysis, ESR, and latex rheumatoid factor
results were obtained at each clinical assessment. Elevations
of SGOT to 3 times that of normal levels, a decrease in the
white cell count to <3,500/mm3, or a decrease in the platelet
count to <1.5 x 10s/mm3was considered cause to withhold
MTX until these values returned to normal, at which time
the same dosage would be restarted. Elevations of SGOT or
Table 1. Characteristics of 29 rheumatoid arthritis patients
undergoing methotrexate therapy at the Albany Medical Center
Age, mean (range)
Disease duration, mean (range)
Prednisone (mean dosage)
DR4 positive
Rheumatoid factor positive
WhitelblacklAm. Ind.*
* Am.
American Indian.
50.2 (21-78)
139.7 (95-384)
21/29 (6.2 mg baseline)
GGT .which were <3 times hormal were reevaluated weekly.
Elevations in SGOT levels that remained 3 times the normal
range for a period of 1 month were considered reason to
terminate the study and perform liver biopsy. Because
patients took MTX on different days of the week, blood
samples were obtained in the period between a few hours
and several days after MTX ingestion.
Twenty-eight patients were typed for HLA-A, B, C,
and D polymorphism in the Paternity Testing Laboratory of
the Department of Surgery; the reagents included antiserum
tested during the 8th Histocompatibility Workshop.
Radiographic assessment. Radiographs were obtained
at baseline, 2 years, and annually thereafter. Radiographic
studiecj were evaluated by a bone radiologist in an open,
unblinlded fashion. The radiologist had no knowledge of the
patients' clinical response to methotrexate, but was aware of
the proper chronologic sequence of the radiographs. Radiographs of hands, wrists, and feet were rated in 5 categories,
including soft tissue swelling, erosions, joint space narrowing, malalignment, and osteoporosis. A patient was rated as
improved (or worsened) if the category of disease seen on
radiographs had changed since the previous reading. In
considering these changes, careful attention was given to
radiologic technique, to ensure that it was identical to
baseline. This involved direct communication between the
bone radiologist and the technician taking the film. Erosions
were rated according to shape, appearance, size, and number present. For a patient to be scored as improved or
worsened in this category, a noticeable change in these
parameters, compared with the previous radiographs, would
Table 2,. Average values (mean t standard deviation) of response variables by months of therapy and no. of patients
Months of therapy (no. of patients)
No. tender joints
No. swollen joints
Morning stiffness (minutes)
Grip strength (mm Hg)
SO-foot walk time (seconds)
Pain evaluation (0-4 scale)
Global evaluation (0-4 scale)
Hemoglobin (gm%)
Erythrocyte sedimentation rate
Platelets, (103/mm3)
White blood cells (IO'/mm3)
15.1 t 8.2
16.8 t 7.4
162 t 173
65 t 45
16.8 t 11.8
12.7 t 11.6
14.4 t 7.8
1 1 3 t 119*
75 t 38*
12.6 2 45*
11.8 t 14.4*
12.0 t 9.2
70 t 83*
93 t 85
13.1 t 7.1*
2.5 t 0.5
2.3 t 0.4
1.6 t 0.7*
1.7 t 0.7'
1.5 t 1.0*
1.5 t 0.9*
1.5 t 0,7*
1.4 t 0.5*
2.5 t 0.5
2.4 t 0.6
12.5 t 1.41
36.2 t 24.7
345 t 132
9.0 t 2.2
1.8 t 0.7*
1.8 -+ 0.7*
12.4 t 1.3
29.6 t 17.0
332 t 92
8.3 t 2.4
1.7 t 0.8*
1.8 t 0.8*
12.9 t 0.9
32.4 2 22.5
306 t 103
8.4 t 2.4
1.4 0.5*
1.4 C 0.5*
13.0 t 1.0*
24.5 t 12.9
324 t 104
8.6 t 2.5
t 9.2*
t 7.3*
t 95*
t 4.8*
5.5 t 5.6*
9.8 t 4.6*
72 t 93*
75 t 43*
10.7 t 2.3*
1.9 2 0.9*
1.6 3 1.0*
1.5 t 1.0*
1.6 t 0.8*
2.0 t 1.0"
1.9 f 0.8*
13.1 2 1.2*
24.3 t 14.2
298 2 80*
8.2 f 2.2
1.6 t 1.0*
1.5 t 0.7*
13.3 t 1.2*
23.4 t 13.0
298 t 80*
7.8 t 1.9
t 6.8
t 54*
Table 2. (continued)
Months of theram (no. of Datients)
No tender joints
No. swollen joints
Morning stiffness (minutes)
Grip strength (mm Hg)
5O-foOn walk time (seconds)
Pain evaluation (0-4 scale)
Global evaluation (0-4 scale)
Hemoglobin (gm%)
Erythrocyte sedimentation rate
Platelets (103hm3)
White blood cells (103/mm3)
6.4 t 6.2*
10.8 t 5.3
54 t 65*
83 f 56*
11.5 t 2.7*
6.0 t 5.8*
8.6 t 5.5*
44 t 44*
86 t 63
13.3 2 9.0
6.4 t 5.5*
10.5 t 4.5*
64 t 88*
75 t 58
13.6 t 8.1*
6.5 t 6.4*
10.0 t 6.9
8 4 t 182
81 t 48
12.3 t 3.1*
5.5 t 6.0
10.4 t 8.2*
57.8 t 66.8
62.6 t 23.2
14.0 t 6.8
5.1 t 6.7
11.2 t 7.9
42.8 t 64.5
75.7 t 38.2
10.3 t 2.0
5.6 t 4.8
10.7 t 7.4
30.8 t 25.7
109.8 t 53.5
9.1 t 1.5
1.7 2 0.8*
1.5 t 0.5*
1.4 t 1.0*
1.3 t 0.9*
1.7 t 0.8*
1.6 t 0.7*
1.8 t 0.9*
1.6 t 0.7*
1.7 2 0.9*
1.4 t 0.5*
1.7 t 0.7*
1.7 t 0.5*
1.8 t 0.8*
1.5 t 0.5*
1.6 t 0.7"
1.5 t 0.5*
13.2 t 1.0*
26.4 t 18.2
311 t 80
7.7 t 2.0
1.5 t 1.0*
1.5 f 0.8*
13.4 t 1.2*
23.6 f 14.8
305 t 82
7.6 t 2.2
1.5 t 0.9*
1.6 t 0.5*
13.1 t 1.1*
21.8 t 14.6
301 t 76*
8.0 t 2.5
1.6 t 1.0*
1.5 t 0.5*
13.4 2 1.0*
24.8 t 18.3
281 2 56*
7.8 ? 2.3
t 0.7*
1.8 2 0.7*
1.3 t 0.5*
12.9 f 1.4
31.8 2 19.1
287 t 80
7.5 & 1.4
* P < 0.05 versus baseline, by paired t-test.
t 0.5*
t 1.3"
t 17.1
t 67*
t 1.8
t 1.3
2 11.7
t 81
t 2.1
be necessary. Soft tissue swelling and joint space narrowing
were compared visually. The angle between the metacarpal
or metatarsal head and the long axis of the proximal phalanx
was compared for determination of malalignment. Osteoporosis was rated by comparing the width of the cortex of the
metacarpal diaphysis with that seen on previous films.
Visual inspection of changes in transverse trabeculation in
trabecular bone and endosteal thinning of the metacarpals
was also used. Followup films were compared with those
taken at baseline, and were scored in each category as
improved, unchanged, or worsened. Eleven patients had
complete films and are included in this report.
Statistical analysis. Statistical comparisons were
made using the appropriate paired r-test, t-test, or comparison of binomial proportions tests. The chi-square test was
used to evaluate HLA data.
The average age of the patients was 50.2 years
(range 21-78), and the average disease duration was
139.7 months (range 95-384). The 29 patients received
an average weekly dose of 12.4 mg (range 5-17.5) oral
metlhotrexate. The average duration of therapy was
29.1 months (range 7-54 months). Patient demographic
characteristics on entry into the study are shown in
Table 1.
Two patients were withdrawn from the study.
One patient, a 46-year-old woman who had received a
total cumulative dosage of 17.5 mg of MTX over a 4week period (7.5 mg/week once, 5 mg/week once, 2.5
mg/week twice), refused to continue treatment after
experiencing moderate nausea. The nausea resolved
shortly after MTX was discontinued, but the patient
diedl from inanition and chronic debilitation 3 months
later. We did not consider her death to be MTX-
L,, ,v
, 9
28 27
-40 -
TIYE(months) 1 3
related. The other patient, a 53-year-old woman, was
withdrawn because of a hypersensitivity pneumonitis
secondary to MTX.
Effects of MTX on disease. Clinical response.
Average values of the response variables for each visit
are presented in Table 2. Improvement was evident in
the number of tender and swollen joints, length of time
of morning stiffness, grip strength, walk time, and
physician and patient evaluation of pain and global
disease. Improvement was usually observed at the
1-month visit and was continued until the 6-month
visit, at which time a plateau was reached. The effects
generally persisted with continued MTX therapy for 30
months. Since only 12 of the patients have been followed longer than 30 months, subsequent averages may
be substantially influenced by a single observation.
During the first 6 months of treatment, the
average number of tender joints decreased from 15.1
to 8.1, the average number of swollen joints decreased
from 16.8 to 10.4, length of morning stiffness decreased from 162 to 75 minutes, grip strength increased from 65 mm Hg to 82 mm Hg, 50-foot walk
time decreased from 16.8 seconds to 1 I .4 seconds, and
physician and patient evaluation of pain and global
disease each improved from approximately 2.4 to 1.4.
These changes were all statistically significant at the
0.05 level. Plots of the average percent change from
baseline for tender and swollen joints and global
disease activity rating are shown in Figures 1 and 2.
Two patients (6.8%) achieved complete remission according to American Rheumatism Association
criteria (18) after 2 years of MTX therapy. Three
patients (10.3%) were noted to have an increase in the
, ,
, ,
21 24
30 33 36
TIME(months) 1 3
28 24
16 12
- 6 0 L I I
, ,
21 24
30 33
16 12
Figure 1. Mean percent change in clinical parameters from baseline through 36 months, in 29 patients taking methotrexate. A, Number of
tender joints. B, Number of swollen joints.
"rt t
04 Patient
TIME (mos.)
01 3 8 9 12 15 18 21 24 21 30 33 36
02928 28 27 26 24 22 21 19 17 16 12 9
Figure 2. Physician and patient evaluation of global arthritis activity. P < 0.001 at 6 months for both values. In all values for both pain
and global activity, statistically significant improvement was maintained through 36 months.
number of rheumatoid nodules while receiving MTX.
This phenomenon was not prospectively evaluated
and became retrospectively evident after 1 and 2 years
of met hotrexate therapy.
Laboratory response. Average values for hemoglobin, platelet count, and ESR are seen in Table 2.
Hemoglobin levels improved from an average of 12.5
gm% at baseline to 13.0 gm% after 6 months of
treatment ( P < 0.05) and a statistically significant
improvement was maintained through the 27-month
visit. Maximum improvement occurred at 24 months,
at which time the average hemoglobin concentration
had improved by 1.1 gm%. The ESR improved from an
average of 36.2 mm/hour at baseline to 24.5 mm/hour
at 6 months ( P = 0.06), but improvement did not ever
achieve statistical significance.
Average platelet counts (Table 2) decreased
from 345,000/mm3 at baseline to 298,000/mm3 at 9
months ( P = 0.044) and had decreased to 280,000/mm3
at the time of the 24-month visit 'I( = 0.001).
'4dverse occurrences. Side effects of methotrexate therapy were noted in 26 of 29 patients (90%) at
some time during therapy. Gastrointestinal toxicity
was most common (18 of 29 patients, 62%). Abdominal
discomfort was reported on 39 occasions, with nausea
noted :!5 times and diarrhea noted 12 times. Oral
ulcerations or mouth soreness without visible ulcer-
ations were noted on 14 separate occasions in 11
patients. Lightheadedness, headaches, or vertigo were
reported on 13 occasions by 6 patients. One patient
had a cyclic, low-grade fever (100-101°F) for 24 hours
after taking MTX; this phenomenon disappeared during the third year of MTX therapy.
Most reactions were mild or moderate. Only 2
were noted by patients as being severe. One occurred
in a 53-year-old nonsmoking woman who had an acute
onset of fever (102°F) and developed cough, dyspnea,
and pulmonary infiltrates after 7 months of 15 mg/week
of MTX therapy. Blood gas levels obtained at the time
of initial presentation showed a Poz of 36 mm Hg on
room air, and an open-lung biopsy showed the classic
features of hypersensitivity pneumonitis. The patient
required intubation and ventilatory support, as well as
parenteral and oral steroids at the equivalent of 60 mg
prednisone in divided dosages. Over a period of 3
weeks, she had a complete recovery without evidence
of clinical or radiologic sequelae, and she refused the
option of reinstituting MTX. Another patient experienced severe diarrhea and abdominal distress which
abated after the methotrexate dosage was decreased.
All other moderate adverse occurrences became mild
and tolerable after reduction of tbe methotrexate dosage. There was a trend for overall toxicity to diminish
with time.
Leukopenia was observed in 6 patients (20.6%),
with total white blood cell counts ranging from
2,800/mm3with 2,100 polymorphonuclear leukocytes
and 470 lymphocytes/mm3 to 4,200/mm3 with 3,300
polymorphonuclear leukocytes and 880 lymphocytes/
mm3. Three of the 6 episodes occurred shortly after a
low-grade fever and a viral syndrome. Three patients
had resolution of leukopenia after methotrexate was
decreased from 15 mg/week to I0 mg/week in 2 patients, and from 7.5 mglweek to 5 mgfweek in another
patient. One patient had to have MTX withheld for a
period of 3 weeks and reinstituted at a lower dosage,
before resolution of the leukopenia occurred. In 2
patients, leukopenia resolved spontaneously after 1
week, with no lowering of the MTX dosage. Leukopenia occurred at intervals from 1 to 29 months of
treatment. No episodes of thrombocytopenia were
Some elevation in SGOT levels above the normal range occurred in 20 patients (68.9%). Seventeen
patients (58.5%) had elevated SGOT levels on more
than one occasion, 9 patients (31%) had SGOT elevations on at least 3 separate measurements. Three
patients had elevations of SGOT on 6 different occa-
sions, 5 patients had twofold elevations in SGOT
levels at some time in their disease course, and 2 other
patients had a transient fourfold and fivefold increase,
respe:ctively. Increases in transaminase levels tended
to be noted randomly at any time during the course of
therapy, and were more commonly seen in the first 48
hours following a MTX dosage. None of the patients
had persistent significant abnormalities on liver function tests that required premature termination of
Changes in methotrexate and prednisone dosage.
Sixteen patients had to have their dosages of
methotrexate lowered during their treatment because
of either gastrointestinal, oral, or central nervous
system toxicity. These included 9 gastrointestinal reactions, 5 oral reactions, and 2 episodes of headache
or lightheadedness. An initial dosage reduction of 2.5
mg was not always enough to alleviate a side effect; 7
patients required a dosage reduction >2.5 mg before
the reaction was either relieved or became tolerable. It
was inecessary to permanently reduce the MTX dosage
to 510 mg/week in 9 patients. One patient was only
able to tolerate 5 mg/week, yet clinical efficacy was
maintained with this dosage. Three other patients had
their methotrexate reduced to 7.5 mg/week with clinical efficacy maintained. Three instances of dosage
reduction occurred after 1 year of therapy. Switching
from the once-a-week regimen to 3 consecutive 12hour intervals on the same day each week was generally not effective in relieving symptoms of toxicity in
the 8 patients who made this change.
Seven patients had their methotrexate dosage
increased during their disease course. Four patients
had the dosage increased from 15 to 17.5 mg/week
because of a declining therapeutic response after an
average of 33 months of treatment. Three of the 4 had
an initial therapeutic response which equalled, or
exceeded, the mean joint count improvements for the
entire study group (Figures IA and B) during their first
year of MTX therapy. No significant improvement has
been observed in this group after several months of
therapy at the higher dosage. Two patients older than
65 had their methotrexate dosage increased from 7.5 to
12.5 mg/week because of an inadequate initial response to the drug. Both patients improved without
any additional toxicity. The dosage of 1 elderly patient
was increased from 7.5 to 12.5 mg/week because of
lack of efficacy; however, after several months of
receiving the higher dosage, there has been no additional improvement in this individual.
Table 3. Specific changes seen in sequential liver biopsy specimens from patients taking methotrexate
2 years
1.46 gm)
3 years
2.2 gm)
4 years
3.0 gm)
* Two patients had progressed from class I to class 11.
t One patient reverted from class I1 to class I . For definitions of all
Roenigk classifications (I-IV), see Patients and Methods.
The mean prednisone dosage in 21 patients was
decreased from 6.2 mg at baseline to 2.3 mg ( P < 0.01).
Liver biopsy results. Sequential liver biopsy
results were grouped by Roenigk class and are seen in
Table 3. Twenty-eight of 29 baseline biopsies were
graded as Roenigk class I, and 1 was graded as class 11.
One additional patient exhibited severe fatty infiltration (class 11) at the time of the baseline biopsy and
was excluded from participating in the study. Several
baseline biopsies exhibited mild fatty changes or round
cell infiltration but still fell within class I. Nineteen
biopsies were performed after patients had 2 years of
MTX therapy (mean total dosage = 1.46 gm); 16 were
graded as class I and 3 were graded as class 11. Two
patients had progressed from class I to class I1 at the
2-year biopsy (Table 3 ) .
Ten biopsies performed after 3 years (mean
total dosage 2.2 gm) showed 8 class I and 2 class I1
specimens. One patient reverted from class I1 to class
I at the 3-year biopsy. Two biopsies were obtained at
4 years; these showed 1 class I specimen and I class 11.
There was no correlation between any abnormalities found on liver function tests and the histologic
appearance of hepatic tissue obtained by liver biopsy.
HLA phenotypes did not correlate with any changes
noted at the time of liver biopsy.
HLA testing. HLA typing showed that 14 of 28
patients (50%) exhibited the DR4 phenotype. There
was no significant correlation between any HLA phenotype and either toxicity or drug efficacy (data not
shown). Five patients exhibited the DR2 phenotype,
which has been previously suggested to correlate with
a favorable response to methotrexate therapy (2). We
were not able to confirm this observation, since 1 of
the 5 patients had a less significant mean clinical
improvement than the entire study group, and another
was among the 4 patients who experienced clinical
deterioration after prolonged therapy. The remaining
Table 4.
Changes seen on radiographs in 11 patients with rheu-
matoid arthritis, after 2 years of methotrexate therapy
Soft tissue swelling
Joint space narrowing
No change
* Radiographic findings were stable in all patients after 2 years.
patients with the DR2 phenotype exhibited percentage
decreases in joint counts similar to the rest of the study
Radiographic evaluation. Eleven patients had
radiographic followup at 2 years, 4 patients had evaluations after 3 years, and 2 had evaluations after 4
years. Results, subdivided by diagnostic category, are
given in Table 4. Seven of 11 patients had improvement in erosions (Figure 3 ) and 4 patients exhibited no
change. Six of 11 patients had improvement in soft
tissue swelling; 5 patients exhibited no change. All
improlvements were noted at the 2-year visit (Table 4),
and all patients’ radiographic findings remained stable
thereafter. There were no deteriorations noted in any
category on radiographs.
In this long-term prospective study of the use of
methotrexate in rheumatoid arthritis patients, we have
confirmed that patients respond quickly, and show
continued improvement for about 6 months, at which
time clinical response reaches a plateau while
methotrexate is continued. Morning stiffness, grip
strength, 50-foot walk time, tender and swollen joints,
patient and physician evaluation of pain and global
disease, hemoglobin levels, and platelet counts all
showed improvement that was statistically significant
at the 0.05 level during the first 6 months of treatment.
The levels of improvement in these variables were
generally maintained for an additional 18 months.
Interpretation of the findings a t the longer term
followup is less clear because of the fewer numbers of
patients followed for 30 months or longer.
Adverse effects were quite common: 26 of 29
patients (90%) experienced toxicity. Gastrointestinal
and oral reactions were observed most frequently: 18
Figure 3. Improvement of erosions seen on radiographs of a rheumatoid arthritis patient taking methotrexate. When compared with baseline films (left, arrows), filling-in and improvement of erosions (right, arrows)
are seen over a 2-year period.
of 29 patients, (62%) and 11 of 29 patients (38%),
respectively. Headache and lightheadedness were reported in 6 of 29 patients (20.6%). It is important to
note, however, that only 2 of 103 (1.9%) reported
incidents of these 3 adverse reactions were rated as
severe by patients. One was an episode of diarrhea and
abdominal distress which abated after lowering of the
weekly methotrexate dosage, and the other was episode of MTX pneumonitis which resulted in eventual
complete recovery after discontinuation of the drug.
All moderate adverse effects also responded to dosage
reduction by becoming either mild and tolerable, or
disappeared. Only 1 patient had to discontinue MTX
because of gastrointestinal, oral, or central nervous
system side effects. Altering the dosage regimen from
once weekly to 3 consecutive 12-hour intervals once a
week was generally not successful in alleviating adverse effects.
Leukopenia was not severe in any patient and
was easily manageable. As with all other adverse
reactions, it occurred at any time during the treatment
and corrected spontaneously, or after dosage
reduction of MTX. Three of 6 patients with leukopenia
reported antecedent fevers and viral-type illnesses
within 1 to 2 weeks of the lowered white blood cell
count. Other researchers have suggested that the
observed effect of a viral infection on bone marrow
precursors could depend on previous insults to stem
cells that might alter their capacity for self-renewal
(19). We can speculate that a combined effect of a
presumed viral infection, superimposed on long-term
methotrexate administration, might have been sufficient to cause the leukopenia in 3 of the 6 cases
The frequency with which dosage reduction
was, required to relieve adverse effects (20 of 29
patients) raises questions about the appropriateness of
our starting dosage of 15 mg/week in adults whose
weight was >I00 lbs and who were <65 years old.
Yet, many instances of dosage reduction occurred
only after several months of therapy, and 3 occurred
after 1 year of therapy. Also, in an attempt to boost a
declining therapeutic effect, 7 patients required an
increase in methotrexate dosage.
It can be determined from this experience that
patiients vary greatly in both their susceptibility to
toxicity and their threshold for therapeutic improvement. HLA studies were not helpful in predicting
either a favorable response or occurrence of side
effects. Clearly, we should not be rigid about imposing
a single dosage level of methotrexate on all patients
being treated for rheumatoid arthritis.
Although there were frequent minor abnormalities of baseline liver biopsies, all but 1 of 29 specimens
fell within the class I measurement of hepatic changes
in psoriatic patients receiving long-term MTX therapy
(9). Sequential biopsies showed little change: only 2 of
19 patients had progressed to class I1 at their 2-year
biopsy. As yet, none have shown deterioration sufficient to shift their Roenigk class in biopsies obtained
after 2 years. Only 1 biopsy actually improved sufficently to change Roenigk class during the course of
the study, although several specimens exhibited either
decreased fatty change or a diminution in the extent of
round cell infiltration at the time of the repeat biopsy.
This is probably due to the significant mean decrease
in prednisone dosage in the study population (P <
0.01) combined with a decrease in alcohol consumption, but an actual improvement in the hepatic irregularities known to occur in RA patients (10-12) cannot
be excluded.
It must be said that the issue of whether routine
liver biopsies should be performed on patients receiving long-term methotrexate therapy for RA is subject
to debate. Our results and the findings of other investigators (13,20-23) appear to indicate that biopsies are
not needed. Still, there remains the problem of the
frightening incidence of cirrhosis in psoriasis patients
with long-term MTX usage (4,2426). It is likely that,
by eliminating in our RA patients the risk factors
associated with liver damage in psoriatic patients, we
have lessened the incidence of this problem. Therefore, a strong case might be made against routine liver
biopsies during the first 2 years of therapy. Beyond
that, there are insufficient data to adequately resolve
the issue.
Our radiologic data were not analyzed statistically. Nevertheless, the apparent lack of radiologic
evidence of disease progression in all patients examined, and improvement of erosive disease in 7 of 11
patients over a 2-year period, would appear to be
striking evidence of a therapeutic benefit, and confirm
a previous observation (27). These data suggest that
methotrexate could be included with gold and
cyclophosphamide as the only disease-modifying
agents reported to improve erosive disease (28-30).
Conversely, 3 patients experienced an acceleration of rheumatoid nodulosis which occurred at the
same time that their joint disease improved and was
noted retrospectively. No significant or sustained improvements in the rheumatoid factor titers were seen
throughout the course of the study. Four patients
required an increase in their MTX dosage after pro1onge:d therapy, to boost a declining therapeutic responaie.
Acquired resistance to methotrexate in the
treatment of malignant disease has been welldocumented in the laboratory (31-33). This phenomenon may explain our observation of a common plateau
of clinical efficacy, with a diminished therapeutic
response, in selected patients after prolonged treatment. Patients could have differing susceptibilities to
these effects. We did observe continued improvement
beyond 6 months of therapy in several patients, with 2
patierits achieving a complete clinical remission after 2
years of treatment.
We were able to confirm the absence of a
correlation between elevations of transaminase levels
and hepatic changes noted in psoriatic patients (4,9),
and we extend this observation to RA patients who
were receiving long-term treatment with methotrexate. The finding of 57 separate episodes of transaminase level elevations in 20 of 29 patients (68.9%)
made us examine the utility of this measurement,
especially in view of a general absence of hepatic
deterioration, as found on liver biopsy. We cannot
even conclude that patients with normal findings on
liver chemistry tests have little likelihood of developing sienificant hepatic problems, since the experience
with psoriasis patients has clearly shown that this is
not so (31,33). A rational approach would perhaps be
to continue monthly monitoring of liver function, in
the hope of detecting the rare case of an acute, severe
hepatic reaction (6). Transaminase levels that are
elevated but are less than several times the normal
range should probably have little role in deciding who
should have a dosage reduction of methotrexate, or in
the determination of who should or should not receive
the drug.
In summary, it can be said that methotrexate is
an eEective drug in the treatment of rheumatoid arthritis which has not responded to standard therapeutic
approaches. Nevertheless, certain questions concerning long-term therapy with the drug still need to be
answered. We must determine the true incidence of
long-term hepatic changes, as well as methotrexate
pneumonitis, and we must define the duration of
treatment which is appropriate and safe. We need to
examine why many patients do not continue to improve after 6 months of therapy and devise strategies
to overcome this previously undescribed phenomenon. Before we can confidently assign methotrexate to
its proper position in the hierarchy of therapeutic
agents used in the treatment of rheumatoid arthritis,
these problems will have to be solved.
The authors wish to thank Drs. Lloyd Lininger,
Richard Rynes, and Lee Bartholomew for advice and review
of the manuscript, Dr. Foster Scott for reviewing the pathology slides, and Donna Clark and Diane Madia for secretarial
assistance in the preparation of this manuscript.
Gubner R, August S, Grusburg V: Therapeutic suppression of tissue reactivity. 11. Effect of aminopterin in
rheumatoid arthritis and psoriasis. Am J Med Sci
221:176-182, 1951
Weinblatt ME, Coblyn JS, Fox DA, Fraser PA,
Holdsworth DE, Glass DN, Trentham DE: Efficacy of
low-dose methotrexate in rheumatoid arthritis. N Engl J
Med 312:818-822, 1985
Williams HJ, Willkens RF, Samuelson C 0 Jr, Alarcon
GS, Guttadauria M, Yarboro C, Polisson RP, Weiner
SR, Luggen MC, Billingsley LM, Dahl SL, Egger MJ,
Reading JC, Ward JR: Comparison of low-dose oral
pulse methotrexate and placebo in the treatment of
rheumatoid arthritis: a controlled clinical trial. Arthritis
Rheum 28:721-730, 1985
Weinstein G, Roenigk HH, Maibach HI, Cosmides J,
Halprin K, Millard M: Psoriasis-liver-methotrexate interactions: a cooperative study. Arch Dermatol 108:3642, 1973
J. Willkens RF, Watson MA: Methotrexate: a perspective
of its use in the treatment of rheumatic diseases. J Lab
Clin Med 100:314-321, 1982
6. Steinsson K, Weinstein A, Korn J, Abeles M: Low dose
methotrexate in rheumatoid arthritis. J Rheumatol
9:860-866, 1982
7. GroEGD, Shenberger KN, Wilke WS, Taylor TH: Low
dose oral methotrexate in rheumatoid arthritis: an uncontrolled trial and review of the literature. Semin
Arthritis Rheum 12:333-347, 1983
8. Wilke WS, Mackenzie AH, Scherbel AL, Groff GD,
Taylor TH: Toxicity from methotrexate may be dose
related (letter). Arthritis Rheum 26:119-120, 1983
9. Roenigk HH, Auerbach R, Maibach HI, Weinstein GD:
Methotrexate guidelines revised. J Am Acad Dermatol
6:145-155, 1982
10. Lefkovits AM, Farrow IJ: Liver in rheumatoid arthritis.
Ann Rheum Dis 14:162-168, 1955
11. Dietrichson 0 , From A, Christoffersen P, Juhl E: Morphological changes in liver biopsies from patients with
rheumatoid arthritis. Scand J Rheumatol 5:65-69, 1976
12. Fernandes L, Sullivan S, McFarlane EG, Wojcicka BM,
Warues TW, Eddleston ALWF, Hamilton EBD, Wil-
hams R: Studies on the frequency and pathogenesis of
liver involvement in rheumatoid arthritis. Ann Rheum
Dis 38501-506, 1979
13. Hoffmeister RT: Methotrexate therapy in rheumatoid
arthritis: 15 years experience. Am J Med 12:69-73, 1983
14. Fitzgerald 0, Hanly J, Molony J, Bresnihan B: Poor
long-term results from low dose methotrexate in rheumatoid arthritis (letter). Arthritis Rheum 27399-600,
15. Zachariae H , Kragballe K, Thestrup-Pedersen F,
ECissmeyer-Nielsen F: HLA antigens in methotrexateinduced liver cirrhosis. Acta Derm Venereol (Stockh)
610: 165-166, 1980
16. Ropes MW, Bennett GA, Cobb S, Jacox R, Jessar RA:
1958 revision of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis 9: 175-176, 1958
17. Steinbrocker 0, Traeger CH, Batterman RC: Therapeutic criteria in rheumatoid arthritis. JAMA 140:659-662,
18. I’inals RS, Masi AT, Larsen RA, The Subcommittee for
Criteria of Remission in Rheumatoid Arthritis of the
American Rheumatism Association Diagnostic and
‘Therapeutic Criteria Committee: Preliminary criteria for
clinical remission in rheumatoid arthritis. Arthritis
liheum 24:130%1315, 1981
19. Young N, Mortimer P: Viruses and bane marrow failure.
]Blood 63:729-737, 1984
20. IMackenzie AH: Liver biopsy findings after methotrexate therapy for rheumatoid arthritis (abstract). J
IRheumatol 1:S73, 1974
21. ’Willkens RF, Clegg DO, Ward JR, Marks CR, Greene
IML, Tolman KG, Roth GJ, Jackson CG, Leonard PA:
]Liver biopsies in patients on low dose pulse methotrexiite for the treatment of rheumatoid arthritis (abstract).
.4rthritis Rheum (suppl) 28:S77, 1985
22. Aponte J, Petrelli M: Hepatic histology following prolonged treatment of rheumatoid arthritis with bolus
~methotrexate(abstract). Arthritis Rheum (suppl) 28:S37,
23. DiBartolomeo AG, Mayes MD, Bathon JM: Methotrex-
83 1
ate in rheumatoid arthritis: a longitudinal study of liver
biopsies (abstract). Arthritis Rheum (suppl) 27:S61, 1984
24. Nyfors A: Liver biopsies from psoriatics related to
methotrexate therapy. 3. Findings in post-methotrexate
liver biopsies from 160 psoriatics. Acta Pathol Microbiol
Immunol Scand [A] 85511-519, 1977
25. Robinson J, Baughman RD, Auerbach R, Cimis RJ:
Methotrexate hepatotoxicity in psoriasis: consideration
of liver biopsies at regular intervals. Arch Dermatol
116:413-415, 1980
26. Roenigk H, Berfeld W, St. Jacques R, Owens F, Hawk
W: Hepatotoxicity of methotrexate in the treatment of
psoriasis. Arch Dermatol 103:250-261, 1971
27. Butler D, Tiliakos N: Low dose cytotoxic drug combination therapy in intractable rheumatoid arthritis (abstract). Arthritis Rheum (suppl) 28:S15, 1985
28. Cooperating Clinics Committee of the American Rheumatism Association: A controlled trial of gold salt therapy in rheumatoid arthritis. Arthritis Rheum 16:353-358,
29. Sigler JW, Bluhm GB, Duncan H , Sharp JT, Ensign DC,
McCrum WR: Gold salts in the treatment of rheumatoid
arthritis: a double-blind study. Ann Intern Med
80:21-26, 1974
30. Cooperating Clinics Committee of the American Rheumatism Association: A controlled trial of cyclophosphamide in rheumatoid arthritis. N Engl J Med 283:
883-889, 1970
31. Hill BT, Baley VBD, White JC, Goldman ID: Characteristics of transport of 4-amino antifolates and folate
compounds by two lines of L51784 lymphoblasts, one
with impaired transport of methotrexate. Cancer Res
39:2440-2446, 1979
32. Fliutoff WF, Essani K: Methotrexate-resistant Chinese
hamster ovary cells contain a dihydrofolate reductase
with an altered activity for methotrexate. Biochemistry
19:4321-4327, 1980
33. Cart GA, Carney DN, Cowan KH, Jolivet J, Bailey BD,
Drake JC, Kao-Shan CS, Minna JD, Chamber BA:
Unstable methotrexate resistance in human small-cell
carcinoma associated with double minute chromosomes.
N Engl J Med 3083199-202, 1983
Без категории
Размер файла
999 Кб
efficacy, terms, long, arthritis, methotrexate, safety, use, therapy, rheumatoid
Пожаловаться на содержимое документа