Long-Term
Methotrexate
By Christopher
Therapy
J. Scully, Cynthia
A retrospective review of methotrexate (MTX) treatment assessed the clinical course in 124 rheumatoid arthritis (RA) patients. After 5 years, 39 (31%) patients continued MTX with clinical benefit. Although patients continuing MTX after 5 years were younger (45 2 13 v 54 ? 12 yrs, P < .OOl) and had a shorter disease duration of RA (9.3 + 8.1 v 14 ? 11 yrs, P < .05) than patients who discontinue the drug, these differences were not considered clinically significant. MTX was discontinued in 20 patients for a lack of clinical benefit, in 21 patients for non-drugrelated reasons, and in 44 patients for suspected adverse drug reactions. The adverse drug reactions requiring permanent discontinuation of MTX were nausea, stomatitis, hair loss, rash, pulmonary reactions, elevated liver enzymes,
D
URING the last decade, short-term, prospective, double-blind, placebo-controlled ” trials have established methotrexate (MTX) as an effective drug for rheumatoid arthritis (RA). Additional investigations have shown MTX comparable with standard second-line agents used to treat RA.Sy Uncontrolled studies have also described clinical improvement with MTX”‘.” with sustained clinical benefit reported after 3 to 4 years of follow-up. These previous long-term reports describe patients receiving MTX for differing treatment intervals. Few patients in these studies have continued the drug as long as 5 years. Long-term studies in which patients are observed for fixed treatment intervals are needed to define efficacy, tolerability. and safety and to further identify the type and frequency of adverse drug reactions associated with the prolonged use of MTX. This report provides a j-year follow-up of 124 RA patients treated with MTX at our institution. The experience with this group after 2 years of therapy has been reported previously.‘” Specific goals of this project were to (1) determine the number of patients continuing MTX with sustained clinical benefit after five years of treatment, (2) determine the reasons for discontinuing MTX, (3) identify potential factors associated with long-term clinical benefit or toxicity, and (4) determine if there are differences in the type and severity of adverse reactions developSemfnars
for Rheumatoid
J. Anderson,
Arthritis
and Grant W. Cannon
hematologic abnormalities, and hepatic fibrosis. At least one adverse drug reaction was reported by 115 (93%) patients receiving MTX, but the majority did not require permanent drug discontinuation. Although the prevalence of adverse reactions increased with longer duration of therapy, no differences existed in the type of reactions reported over 5 years of treatment. There were no risk factors identified that were clearly associated with the development of toxicity. Long-term therapy was primarily limited by adverse reactions rather than loss of efficacy. Copyright
o 1991 by W.B. Saunders
INDEX WORDS: Rheumatoid otrexate; cytotoxic drugs.
Company
arthritis;
meth-
ing early in the course of therapy compared with those occurring late in therapy. The literature on long-term therapy of RA with MTX also is reviewed. METHODS
All patients at the University of Utah Medical Center and the Salt Lake City Veterans Affairs Medical Center meeting ARA criteria for classic or definite RA” who started MTX prior to December 31, 1983 were identified. Each patient was evaluated for the 5 years after the start of MTX, regardless of their present treatment status. Methods of data collection, determination of clinical response, laboratory testing, and previous antirheumatic drug therapies have been described previously.‘” Patients were classified at their 5-year anniversaries as either “on MTX” or “off MTX.” Temporary interruption of ther-
From the Division of Rheumatoioa,
Universig
of Utah arid
the VA Medical Center, Salt Lake City, liT. Supported
by the Department
of Veterans Affain
Medical
Research Program and the Nora Eccles Treadwell Foundation.
Christopher J. Scully, MD: Fellow in Rheumatolo~; Qnthia J. Anderson, MD: Fellow in Rheumatology; Grant W. Cannon, MD: Associate Chief of Staff’for Education and Assistant Professor of Medicine. Address
reprint requests to G. W. Cannon,
Rheumatology,
MD, Division of
50 Medical Dr, Salt Lake City, UT X4132.
Copyright 0 1991 by W.B. Saunder.s Compurzy 0049-017219112005-0005$5.OOlO
,n Art/?&s and Rheumatism, Vol 20, No 5 (April), 1991: pp 317.331
317
318
apy was often necessary without requiring permanent discontinuation of the drug. An “interrupted” course was defined as suspension of MTX therapy for greater than 3 consecutive months and/or a cumulative time off drug of greater than 6 months during the 5-year follow-up period. Patients continuing MTX at 5 years (on MTX) who did not have an interrupted course were considered to have received a “continuous” course. In each patient receiving at least 12 weeks of MTX treatment, the attending physician assessed the outcome as either clinical improvement (MTX responder) or no clinical improvement (MTX failure). This judgment was based on the physician’s global assessment of the severity of morning stiffness, joint pain, and joint swelling as documented in the medical record. The degree of improvement or failure was not quantified. Adverse drug reactions were recorded and the physician’s response to the event was classified as MTX discontinued, MTX dosage changed (temporarily suspended or MTX dose decreased), or MTX dosage unchanged. Reasons for discontinuing MTX were recorded as lack of clinical response (clinical failure), drug toxicity, or non-drug-related (concurrent medical illness, patient request, lack of cooperation, or lost to follow-up). Complete blood count (CBC), liver enzymesaspartate aminotransferase (AST), alanine aminotransferase (ALT), y glutamyl transferase (GGTP), alkaline phosphatase (ALP), and lactate dehydrogenase (LDH) were obtained at 4 to 8 week intervals. Liver enzymes and CBC were recorded as abnormal if any test was outside the laboratory reference range. Blood urea nitrogen, and serum creatinine IeveIs were obtained every 2 to 6 months. Initial laboratory data for patients “on MTX” and “off MTX” at the 5-year anniversary were compared searching for factors present at the time MTX therapy was initiated that could be associated with long-term benefit. Similarly, comparisons were made for patients with and without adverse drug reactions to determine risk factors associated with the development of toxicity. Initial and 5-year laboratory data for patients “on
SCULLY, ANDERSON,
AND CANNON
MTX” were compared to determine if changes developed during MTX treatment. Percutaneous liver biopsy was recommended to all patients continuing MTX longer than 2 years and at 2-year intervals thereafter. Biopsies were graded, using a modification of the Roenigk classification as described by Tolman et al:34 grade I, normal; grade II, nuclear variation and/or fatty infiltration; grade III, necrosis and/or inflammation; grade IV, fibrosis; grade V. cirrhosis. Statistics Unless otherwise stated, data are presented as mean ? standard deviation. Comparisons between groups was by student t-test or x’ analysis as appropriate. Comparisons within groups of initial and final laboratory data was by paired t-test. A P-value of < .05 was considered statistically significant. RESULTS
One hundred twenty-four patients met inclusion criteria. The mean age was 5 1 years (range 20 to 76) and mean duration of RA at MTX initiation was 12 years (range 1 to 55). Fifty-one patients (41%) were men, and 73 (59%) were women. At 5 years, 39 patients (31%) continued to receive MTX with sustained clinical benefit (Fig 1). Thirty-two patients (82%) received a “continuous” course, whereas seven patients (18%) had an “interrupted” course with a total time “off MTX” ranging from 5 to 10 months. Eighty-five patients (69%) discontinued MTX before completing 5 years of therapy (Fig 2). Of these, 20 patients discontinued MTX for lack of clinical benefit, 13 of whom never achieved a clinical response after at least 12 weeks of treatment with at least 7.5 mg MTX per week. Seven patients who experienced an initial clinical response subsequently discontinued the drug after worsening of RA. Forty-four patients (35%) discontinued MTX for suspected adverse drug reactions. Twenty-one patients (18%) discontinued MTX for non-drug-related reasons, including 9 with concurrent medical illness, 7 for personal reasons (expense, fear of toxicity or liver biopsy, or unspecified), and 5 were lost to follow-up.
LONG-TERM
319
MTX THERAPY FOR RA
AJr
0
;
1
A
b
k
Years on Methotrexate Fig 1:
Numbers of patient continuing to receive methotrexate
One hundred fifteen patients (93%) reported one or more adverse reactions (Table 1). The majority of adverse reactions were minor, transient, and did not require a change in therapy. Nausea was the most frequent symptomatic side effect, reported by 49 patients (39%), followed by stomatitis in 36 patients (29%), hair loss in 26 patients (21%) and rash in 13 patients (10%). Pulmonary symptoms occurred in 11 patients (9%) and resulted in discontinuation of the drug for 8 patients. Clinical information on 9 of these patients with pulmonary reactions are detailed in two earlier reports.“5..‘h Elevated liver enzymes occurred in 87 patients (70%) and resulted in the discontinuation of MTX for 9 patients. Eight of these 9 patients discontinued the drug in the first year of therapy and the other patient after 18 months of therapy. One additional patient discontinued MTX after simultaneously experiencing stomatitis, liver enzyme abnormalities (ALT, 73; GGTP, 197) and a disease flare. Nine patients developed leukopenia (range 2.6 to 3.9 x lO”il~,L), and three patients experienced thrombocytopenia (range 83 to 119 x lo31
over 5 years of treatment.
l.r,L). All abnormal blood counts resolved after temporary or permanent withdrawal of MTX. Fifty-seven liver biopsies were obtained from 40 patients (Fig 3). Fifteen patients underwent two biopsies and one patient underwent three biopsies. The mean dose of MTX was 1.3 2 0.6 g with a mean duration of treatment of 2.7 ? 0.9 yrs at the time of the first liver biopsy. At the time of the second biopsy, the mean dose of MTX was 1.7 * 0.4 g with a mean duration of treatment of 3.9 2 0.6 yrs. The patients undergoing a third liver biopsy had received 2.6 g of MTX over 4.8 yrs, at the time of the last biopsy. Ten biopsies were grade I biopsies, 21 grade II, nine grade III, and 17 grade IV. No patient had cirrhosis (grade V). For those with multiple biopsies, histologic grade remained unchanged in 10, improved in 4, and worsened in 2. Two patients discontinued MTX as a direct result of histologic findings, both having grade IV changes on two occasions. MTX dose was lowered in one patient because of fibrosis, but in all other patients the MTX dose was unchanged. No correlation was found between liver histology and abnormal liver enzymes.
SCULLY, ANDERSON,
1
2
3
AND CANNON
4
5
Year During Which Methotrexate Was Discontinued Fig 2:
Reasons
for discontinuing
methotrexate
1. Adverse
Drug Reactions
of treatment.
Mean weight (kg) and male:female ratio were not significantly different. Initial laboratory values and mean initial prednisone doses were similar for patients in both groups. A comparison of laboratory and clinical parameters, previous clinical response and adverse reactions with second-line drug therapy, and alcohol consumption in patients with and without adverse drug reactions requiring that
Comparison of the “on MTX” group with those “off MTX” shows mean age and disease duration to be statistically significantly lower for patients continuing MTX at 5 years than for those discontinuing the drug (mean age 45 2 13 yrs, v 54 f 11 yrs, P < .OOl; mean disease duration 9.3 + 8.1 yrs v 14 2 11 yrs, P < .05); however, these differences were not considered to be a clinically significant difference (Table 2). Table
over 5 years
Associated
Alteration
With
MTX
in 124 RA Patients
of MTX Dose
Total Number of
Dose
Dose
MTX
Patients With the
Unchanged
Changed
Discontinued
Adverse Reaction
Clinical Nausea
31 (63%)*
11 (22%)
7 (15%)
49
Stomatitis
15 (42%)
12 (33%)
9 (25%)
36
Hair Loss
21 (81%)
1 (4%)
4 (15%)
26
Rash
5 (38%)
5 (38%)
3 (24%)
13
Pulmonary
0 (0%)
3 (27%)
8 (73%)
11
Laboratory Elevated LFT
63 (72%)
9 (11%)
a7
Hematologic
11 (52%)
8 (38%)
2 (10%)
21
Hepatic Fibrosis
14 (82%)
1 (6%)
2 (12%)
17
15 (17%)
*Number in parentheses is percent of patients with the specific adverse reaction. Abbreviation: LFT. liver function
tests.
321
LONG-TERM MTX THERAPY FOR RA
Grade
First Biopsy
Second Biopsy
Third Biopsy
Total
xx xx I
IO
v
0
Fig 3: Results of sequential discontinued.
liver biopsies. Parentheses
MTX be discontinued showed no statistically significant differences between the two groups. Comparisons of initial and 5-year follow-up data in patients continuing MTX at 5 years (on MTX) showed a statistically significant reduction in mean daily prednisone dose, 6.3 & 5.0 mg to 4.0 + 3.3 mg (P < .05). The only statistically significant change in laboratory values for this group was an increase in the mean corpuscular volume (MCV), 84 +- 7 to 92 IT 7 (P < .OOl), which was not considered clinically significant. DISCUSSION
AND LITERATURE REVIEW
Our study is a retrospective review of 124 RA patients followed for a fixed five-year interval following the initiation of MTX. To compare
mark patients
in whom
methotrexate
was
our results with previous reports, an understanding of the strengths and limitations of the different methodologies is necessary. Our method has the advantage of evaluating a large cohort of patients over a ftxed time period and allows calculation of the incidence of adverse drug reactions. Other retrospective reviews have not followed patients for a fixed time interval, but rather report their experience to a fixed date and, thus, for variable treatment intervals. The retrospective design lacks prospective criteria to quantify the degree of clinical response and to ensure that the management of potential adverse reactions was uniform. Prospective studies have uniform criteria for clinical response and management of adverse drug effects but are often limited by the small number of patients
322
SCULLY, ANDERSON, AND CANNON
Table 2. Patient Demographics Dosing in Patients
and MTX
on and off MTX
patients (54%, range 43% to 82%) continuing the drug. At 4 years, three studies, one prospective’? and two retrospective reviews (including the present results),2’ report 65 of 163 patients (40%, range 37% to 50%) continuing the drug. After 5 years, 31% of our patients continued to receive MTX with sustained clinical benefit, the majority of these patients received a continuous course of therapy. Alarcon et al,” in a review of 152 patients, calculated the probability of continuing MTX for 5 years to be 49%, although for a subgroup of 72 patients in whom the drug was started prior to January 1984, the probability of continuing MTX for 5 years was only 38%. This difference was believed to reflect a possible change over time with respect to greater physician experience and familiarity with the drug and its side effects. Hoffmeister14 reported 22 patients continuing MTX for 5 to 15 years, and Kremer and Lee,‘” in follow-up of a prospective study of 29 patients, reported 7 patients continuing MTX for at least 60 months. Other studies published have included even fewer patients. I2 I3 15-’18,23,24.27.29.31.32 Studies with a reported mean MTX treatment duration longer than 18 months show that the proportion of patients continuing MTX at the different time intervals is higher in the
at 5 Years
MTX Status at 5 Years
Age (vrs)
On
Off
P
45 * 13
54% 12
Methotrexate
By Christopher
Therapy
J. Scully, Cynthia
A retrospective review of methotrexate (MTX) treatment assessed the clinical course in 124 rheumatoid arthritis (RA) patients. After 5 years, 39 (31%) patients continued MTX with clinical benefit. Although patients continuing MTX after 5 years were younger (45 2 13 v 54 ? 12 yrs, P < .OOl) and had a shorter disease duration of RA (9.3 + 8.1 v 14 ? 11 yrs, P < .05) than patients who discontinue the drug, these differences were not considered clinically significant. MTX was discontinued in 20 patients for a lack of clinical benefit, in 21 patients for non-drugrelated reasons, and in 44 patients for suspected adverse drug reactions. The adverse drug reactions requiring permanent discontinuation of MTX were nausea, stomatitis, hair loss, rash, pulmonary reactions, elevated liver enzymes,
D
URING the last decade, short-term, prospective, double-blind, placebo-controlled ” trials have established methotrexate (MTX) as an effective drug for rheumatoid arthritis (RA). Additional investigations have shown MTX comparable with standard second-line agents used to treat RA.Sy Uncontrolled studies have also described clinical improvement with MTX”‘.” with sustained clinical benefit reported after 3 to 4 years of follow-up. These previous long-term reports describe patients receiving MTX for differing treatment intervals. Few patients in these studies have continued the drug as long as 5 years. Long-term studies in which patients are observed for fixed treatment intervals are needed to define efficacy, tolerability. and safety and to further identify the type and frequency of adverse drug reactions associated with the prolonged use of MTX. This report provides a j-year follow-up of 124 RA patients treated with MTX at our institution. The experience with this group after 2 years of therapy has been reported previously.‘” Specific goals of this project were to (1) determine the number of patients continuing MTX with sustained clinical benefit after five years of treatment, (2) determine the reasons for discontinuing MTX, (3) identify potential factors associated with long-term clinical benefit or toxicity, and (4) determine if there are differences in the type and severity of adverse reactions developSemfnars
for Rheumatoid
J. Anderson,
Arthritis
and Grant W. Cannon
hematologic abnormalities, and hepatic fibrosis. At least one adverse drug reaction was reported by 115 (93%) patients receiving MTX, but the majority did not require permanent drug discontinuation. Although the prevalence of adverse reactions increased with longer duration of therapy, no differences existed in the type of reactions reported over 5 years of treatment. There were no risk factors identified that were clearly associated with the development of toxicity. Long-term therapy was primarily limited by adverse reactions rather than loss of efficacy. Copyright
o 1991 by W.B. Saunders
INDEX WORDS: Rheumatoid otrexate; cytotoxic drugs.
Company
arthritis;
meth-
ing early in the course of therapy compared with those occurring late in therapy. The literature on long-term therapy of RA with MTX also is reviewed. METHODS
All patients at the University of Utah Medical Center and the Salt Lake City Veterans Affairs Medical Center meeting ARA criteria for classic or definite RA” who started MTX prior to December 31, 1983 were identified. Each patient was evaluated for the 5 years after the start of MTX, regardless of their present treatment status. Methods of data collection, determination of clinical response, laboratory testing, and previous antirheumatic drug therapies have been described previously.‘” Patients were classified at their 5-year anniversaries as either “on MTX” or “off MTX.” Temporary interruption of ther-
From the Division of Rheumatoioa,
Universig
of Utah arid
the VA Medical Center, Salt Lake City, liT. Supported
by the Department
of Veterans Affain
Medical
Research Program and the Nora Eccles Treadwell Foundation.
Christopher J. Scully, MD: Fellow in Rheumatolo~; Qnthia J. Anderson, MD: Fellow in Rheumatology; Grant W. Cannon, MD: Associate Chief of Staff’for Education and Assistant Professor of Medicine. Address
reprint requests to G. W. Cannon,
Rheumatology,
MD, Division of
50 Medical Dr, Salt Lake City, UT X4132.
Copyright 0 1991 by W.B. Saunder.s Compurzy 0049-017219112005-0005$5.OOlO
,n Art/?&s and Rheumatism, Vol 20, No 5 (April), 1991: pp 317.331
317
318
apy was often necessary without requiring permanent discontinuation of the drug. An “interrupted” course was defined as suspension of MTX therapy for greater than 3 consecutive months and/or a cumulative time off drug of greater than 6 months during the 5-year follow-up period. Patients continuing MTX at 5 years (on MTX) who did not have an interrupted course were considered to have received a “continuous” course. In each patient receiving at least 12 weeks of MTX treatment, the attending physician assessed the outcome as either clinical improvement (MTX responder) or no clinical improvement (MTX failure). This judgment was based on the physician’s global assessment of the severity of morning stiffness, joint pain, and joint swelling as documented in the medical record. The degree of improvement or failure was not quantified. Adverse drug reactions were recorded and the physician’s response to the event was classified as MTX discontinued, MTX dosage changed (temporarily suspended or MTX dose decreased), or MTX dosage unchanged. Reasons for discontinuing MTX were recorded as lack of clinical response (clinical failure), drug toxicity, or non-drug-related (concurrent medical illness, patient request, lack of cooperation, or lost to follow-up). Complete blood count (CBC), liver enzymesaspartate aminotransferase (AST), alanine aminotransferase (ALT), y glutamyl transferase (GGTP), alkaline phosphatase (ALP), and lactate dehydrogenase (LDH) were obtained at 4 to 8 week intervals. Liver enzymes and CBC were recorded as abnormal if any test was outside the laboratory reference range. Blood urea nitrogen, and serum creatinine IeveIs were obtained every 2 to 6 months. Initial laboratory data for patients “on MTX” and “off MTX” at the 5-year anniversary were compared searching for factors present at the time MTX therapy was initiated that could be associated with long-term benefit. Similarly, comparisons were made for patients with and without adverse drug reactions to determine risk factors associated with the development of toxicity. Initial and 5-year laboratory data for patients “on
SCULLY, ANDERSON,
AND CANNON
MTX” were compared to determine if changes developed during MTX treatment. Percutaneous liver biopsy was recommended to all patients continuing MTX longer than 2 years and at 2-year intervals thereafter. Biopsies were graded, using a modification of the Roenigk classification as described by Tolman et al:34 grade I, normal; grade II, nuclear variation and/or fatty infiltration; grade III, necrosis and/or inflammation; grade IV, fibrosis; grade V. cirrhosis. Statistics Unless otherwise stated, data are presented as mean ? standard deviation. Comparisons between groups was by student t-test or x’ analysis as appropriate. Comparisons within groups of initial and final laboratory data was by paired t-test. A P-value of < .05 was considered statistically significant. RESULTS
One hundred twenty-four patients met inclusion criteria. The mean age was 5 1 years (range 20 to 76) and mean duration of RA at MTX initiation was 12 years (range 1 to 55). Fifty-one patients (41%) were men, and 73 (59%) were women. At 5 years, 39 patients (31%) continued to receive MTX with sustained clinical benefit (Fig 1). Thirty-two patients (82%) received a “continuous” course, whereas seven patients (18%) had an “interrupted” course with a total time “off MTX” ranging from 5 to 10 months. Eighty-five patients (69%) discontinued MTX before completing 5 years of therapy (Fig 2). Of these, 20 patients discontinued MTX for lack of clinical benefit, 13 of whom never achieved a clinical response after at least 12 weeks of treatment with at least 7.5 mg MTX per week. Seven patients who experienced an initial clinical response subsequently discontinued the drug after worsening of RA. Forty-four patients (35%) discontinued MTX for suspected adverse drug reactions. Twenty-one patients (18%) discontinued MTX for non-drug-related reasons, including 9 with concurrent medical illness, 7 for personal reasons (expense, fear of toxicity or liver biopsy, or unspecified), and 5 were lost to follow-up.
LONG-TERM
319
MTX THERAPY FOR RA
AJr
0
;
1
A
b
k
Years on Methotrexate Fig 1:
Numbers of patient continuing to receive methotrexate
One hundred fifteen patients (93%) reported one or more adverse reactions (Table 1). The majority of adverse reactions were minor, transient, and did not require a change in therapy. Nausea was the most frequent symptomatic side effect, reported by 49 patients (39%), followed by stomatitis in 36 patients (29%), hair loss in 26 patients (21%) and rash in 13 patients (10%). Pulmonary symptoms occurred in 11 patients (9%) and resulted in discontinuation of the drug for 8 patients. Clinical information on 9 of these patients with pulmonary reactions are detailed in two earlier reports.“5..‘h Elevated liver enzymes occurred in 87 patients (70%) and resulted in the discontinuation of MTX for 9 patients. Eight of these 9 patients discontinued the drug in the first year of therapy and the other patient after 18 months of therapy. One additional patient discontinued MTX after simultaneously experiencing stomatitis, liver enzyme abnormalities (ALT, 73; GGTP, 197) and a disease flare. Nine patients developed leukopenia (range 2.6 to 3.9 x lO”il~,L), and three patients experienced thrombocytopenia (range 83 to 119 x lo31
over 5 years of treatment.
l.r,L). All abnormal blood counts resolved after temporary or permanent withdrawal of MTX. Fifty-seven liver biopsies were obtained from 40 patients (Fig 3). Fifteen patients underwent two biopsies and one patient underwent three biopsies. The mean dose of MTX was 1.3 2 0.6 g with a mean duration of treatment of 2.7 ? 0.9 yrs at the time of the first liver biopsy. At the time of the second biopsy, the mean dose of MTX was 1.7 * 0.4 g with a mean duration of treatment of 3.9 2 0.6 yrs. The patients undergoing a third liver biopsy had received 2.6 g of MTX over 4.8 yrs, at the time of the last biopsy. Ten biopsies were grade I biopsies, 21 grade II, nine grade III, and 17 grade IV. No patient had cirrhosis (grade V). For those with multiple biopsies, histologic grade remained unchanged in 10, improved in 4, and worsened in 2. Two patients discontinued MTX as a direct result of histologic findings, both having grade IV changes on two occasions. MTX dose was lowered in one patient because of fibrosis, but in all other patients the MTX dose was unchanged. No correlation was found between liver histology and abnormal liver enzymes.
SCULLY, ANDERSON,
1
2
3
AND CANNON
4
5
Year During Which Methotrexate Was Discontinued Fig 2:
Reasons
for discontinuing
methotrexate
1. Adverse
Drug Reactions
of treatment.
Mean weight (kg) and male:female ratio were not significantly different. Initial laboratory values and mean initial prednisone doses were similar for patients in both groups. A comparison of laboratory and clinical parameters, previous clinical response and adverse reactions with second-line drug therapy, and alcohol consumption in patients with and without adverse drug reactions requiring that
Comparison of the “on MTX” group with those “off MTX” shows mean age and disease duration to be statistically significantly lower for patients continuing MTX at 5 years than for those discontinuing the drug (mean age 45 2 13 yrs, v 54 f 11 yrs, P < .OOl; mean disease duration 9.3 + 8.1 yrs v 14 2 11 yrs, P < .05); however, these differences were not considered to be a clinically significant difference (Table 2). Table
over 5 years
Associated
Alteration
With
MTX
in 124 RA Patients
of MTX Dose
Total Number of
Dose
Dose
MTX
Patients With the
Unchanged
Changed
Discontinued
Adverse Reaction
Clinical Nausea
31 (63%)*
11 (22%)
7 (15%)
49
Stomatitis
15 (42%)
12 (33%)
9 (25%)
36
Hair Loss
21 (81%)
1 (4%)
4 (15%)
26
Rash
5 (38%)
5 (38%)
3 (24%)
13
Pulmonary
0 (0%)
3 (27%)
8 (73%)
11
Laboratory Elevated LFT
63 (72%)
9 (11%)
a7
Hematologic
11 (52%)
8 (38%)
2 (10%)
21
Hepatic Fibrosis
14 (82%)
1 (6%)
2 (12%)
17
15 (17%)
*Number in parentheses is percent of patients with the specific adverse reaction. Abbreviation: LFT. liver function
tests.
321
LONG-TERM MTX THERAPY FOR RA
Grade
First Biopsy
Second Biopsy
Third Biopsy
Total
xx xx I
IO
v
0
Fig 3: Results of sequential discontinued.
liver biopsies. Parentheses
MTX be discontinued showed no statistically significant differences between the two groups. Comparisons of initial and 5-year follow-up data in patients continuing MTX at 5 years (on MTX) showed a statistically significant reduction in mean daily prednisone dose, 6.3 & 5.0 mg to 4.0 + 3.3 mg (P < .05). The only statistically significant change in laboratory values for this group was an increase in the mean corpuscular volume (MCV), 84 +- 7 to 92 IT 7 (P < .OOl), which was not considered clinically significant. DISCUSSION
AND LITERATURE REVIEW
Our study is a retrospective review of 124 RA patients followed for a fixed five-year interval following the initiation of MTX. To compare
mark patients
in whom
methotrexate
was
our results with previous reports, an understanding of the strengths and limitations of the different methodologies is necessary. Our method has the advantage of evaluating a large cohort of patients over a ftxed time period and allows calculation of the incidence of adverse drug reactions. Other retrospective reviews have not followed patients for a fixed time interval, but rather report their experience to a fixed date and, thus, for variable treatment intervals. The retrospective design lacks prospective criteria to quantify the degree of clinical response and to ensure that the management of potential adverse reactions was uniform. Prospective studies have uniform criteria for clinical response and management of adverse drug effects but are often limited by the small number of patients
322
SCULLY, ANDERSON, AND CANNON
Table 2. Patient Demographics Dosing in Patients
and MTX
on and off MTX
patients (54%, range 43% to 82%) continuing the drug. At 4 years, three studies, one prospective’? and two retrospective reviews (including the present results),2’ report 65 of 163 patients (40%, range 37% to 50%) continuing the drug. After 5 years, 31% of our patients continued to receive MTX with sustained clinical benefit, the majority of these patients received a continuous course of therapy. Alarcon et al,” in a review of 152 patients, calculated the probability of continuing MTX for 5 years to be 49%, although for a subgroup of 72 patients in whom the drug was started prior to January 1984, the probability of continuing MTX for 5 years was only 38%. This difference was believed to reflect a possible change over time with respect to greater physician experience and familiarity with the drug and its side effects. Hoffmeister14 reported 22 patients continuing MTX for 5 to 15 years, and Kremer and Lee,‘” in follow-up of a prospective study of 29 patients, reported 7 patients continuing MTX for at least 60 months. Other studies published have included even fewer patients. I2 I3 15-’18,23,24.27.29.31.32 Studies with a reported mean MTX treatment duration longer than 18 months show that the proportion of patients continuing MTX at the different time intervals is higher in the
at 5 Years
MTX Status at 5 Years
Age (vrs)
On
Off
P
45 * 13
54% 12
