`
`ON THE TOXICITY OF LOW-DOSE METHOTREXATE
`IN PATIENTS WITH ‘RHEUMATOID ARTHRITIS
`
`' SARAH L. MORGAN. JOSEPH E. BAGGOTP. WILLIAM H. VAUGHN, PEGGY K. YOUNG.
`JANET V. AUSTIN. CARLOS L. KRUMDIECK. and GRACIELA S. ALARCON
`
`Thirty-two patients with rheumatoid arthritis
`completed a 24-week, placebo-controlled, double-blind
`trial of follc acid (FA) supplementation during low-done
`methotrenate (MTX) therapy. Adminutration oi the
`daily FA supplement significantly lowered toxicity
`scorer without alerting eflcaey, as measured hy Joint
`counts, joint indlces, and patient and physician evalua-
`tion of discs: activity. Fiileen patients experienced
`some sort of toxicity; 67% were in the placebo group,
`and 33% were in the FA supplement group. Four
`patlentrintheplaoehogrouphadtoxicltyleveisseriour
`enough to require discontinuation ofthe MTX, while no
`patlts in the FA supplement group discontinued MTX
`because oiroxiclty. Low-norrnailnltialplasmn and red
`blood cell folate levels were predictive oihrture toxicity
`with MT)! therapy. We conclude that a daily supple-
`mentotl rngofFAdurlnglow-doseM’l'Xtherapy
`(median dose 7.5 mglweek [164 nmoieal) is useiui In
`lessening toxicity without altering elicacy during the
`tint 6 montlu of treatment.
`
`From the Department ofriutrition Sciences and the Divi-
`sion of Clinical Immunology and Rheumatology. Department of
`Medicine. ‘i'he University of Alabama at Bh-tnlmham.
`Supported by NIH grants SPO!-CA-23103-lo and SP60-
`AR-206l4. Dr. Moran is the recipient of a Future Nutrition
`Leader’: Award from the international Life Sciences institute!
`Nutrition Foundation (I986-I988).
`Sarah L. Morten. MD. RD. FACP: Joseph E. Bassott.
`PhD; William H. Vaughn. 38: Peggy K. Young, CMA; Janet V.
`Austin. BS; Carlos L. Kttundleck. MD. PhD: Graciela S. Aiarcdn.
`MD. MPH, PACP.
`Address reprint requests to Sarah 1.. Morgan. MD. RD.
`FACP. the University of Alabama at Birmingham. Department of
`Nutrition Sciences. Webb Buildin3—2tl2. UAB Station, Birming-
`ham, Al. 35294.
`Submitted for publication May 26. I989; eeeeptedinrevised
`form August 2!. I989.
`
`Arthritis and llharmathm, Vol. 3, No. iuonuery I990)
`
`The use of the folic acid (PA) antagonist, amin-
`opterin, for the treatment of rheumatoid arthritis (RA)
`was iitst reported by Gubner et al in 1951 (I). Numer-
`ous double-blind. placebo-controlled trials have since
`established the eilicacy of N-l0-rnethylantiuopterin
`(methotrcxate; MTX) in the treatment of RA (2-7). A
`meta-analysis of numerous studies has shown that
`patients receiving MTX for RA have a 26% greater
`improvement in their joint counts and a 39% greater
`improvement in pain scores than do control patients
`receiving nonsteroidal antiinflamrnatory drugs
`(NSAIDs) with or without prednisone. The major
`factor
`MTX treatment scents to be its toiicity
`(9,l0). Toxic manifestations, such as nausea, stomati-
`tis. abnormal liver function. cytopenia, and pulmonary
`toxicity. have generally been reported in 30-60% of
`patients (7-10). and in 1 study, 90% of patients expe-
`rienced toxicity (6).
`The folate status of patients before and during
`MTX therapy has received little attention. even
`though some clinical manifestations of folate defi-
`ciency. such as cytopenia, anorexia, stomatitis. and
`gastrointestinal (GI) intolerance (I l,l2), are also ob-
`served as toxic reactions during low-dose MTX treat-
`ment for RA. it was therefore postulated that the
`administration of FA would be useful in reducing toxic
`manifestations that occur during long-term treatment
`with low-dose MTX for RA. We report herein the
`results of a double-blind. placebo-controlled study
`designed to test this hypothesis.
`
`PATIENTS AND METHODS
`
`Patients. Thirty-nine patients with RA that fulfilled
`the American Rheumatism Association 1958 criteria for
`
`Kopie von subito e.V.. gelielert fur Maiwald Palentanwalts Gmbl-i (COMO6X0O702)
`
`Lilly Ex. 2084
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2084
`Sandoz v. Lilly IPR2016-00318
`
`
`
`10
`
`MORGAN ET AL
`
`Table l. Baseline clinical and demographic characteristics of the
`rhetmlaloid arthritis patients studied’
`
`
`
`Polate group
`(II 9 16)
`
`Placebo group
`(II - to)
`
`sulfasalazine. or hydroxycloroquine treatment had been
`discontinued for at least to days.
`During the study. each patient remained under the
`care of his or her rheurnatologist. abstained from alcohol.
`and continued to receive stable doses of aspirin and/or
`NSAlDs. For those taking prcdnisone at study entry, the
`dosage was kept stable. not to exceed lo mglday. It was
`required that both male and female patients either be prac-
`ticing contraception or have no reproductive potential.
`Study design. The study was a double-blind, placebo-
`controllcd trial of 24 weeks duration. Patients were assigned
`to receive folate (1 mg (2.2. umolcsllday) or placebo. The 2
`groups were matched. by the study statistician. on the basis
`of sex, previous use of folatc-containing vitamins. rheuma-
`toid factor (RF) serology, and age. identical FA- and placebo-
`containing capsules were prepared by the lnvastigational
`Drug Service of the University of Alabama Hospital. using
`capsules provided.by Capsugel (Warner Lambert. Green-
`wood, SC). Folic acid ipteroylglutamic acid) was obtained
`from Lederlc Laboratories (Pearl River, NY). Spectropho-
`tornetric analysis indicated that
`the mean : SD tolate
`content was L03 : 0.l6 mg per capsule.
`Folate-containing vitamin preparations. it previously
`used. were stopped {or the duration of the study. Oral MTX
`was begun at a dosage of 2.5-7.5 mg per weelt. and was
`increased in 2.5-mg increments at
`the discretion of the
`treating rheurnatologist. The weekly dosage of MTX did not
`exceed l5 mg, and the median dosage for both groups was
`7.5 myweelt. The MTX tablets were generally ingested in
`equal numbers on 3 consecutive occasions at l7»honr inter-
`vals, always beginning on the same day of the week. Neither
`the investigators. the patients. nor the treating rheumatolo-
`gists were aware of the plaeebolfolic acid capsule assign-
`ments until the study was completed. Patients were with-
`drawn from MTX therapy at the discretion of the treating
`rheumatologist.
`Patients were evaluated immediately prior to initia-
`tion of MTX (visit 1). and alter approximately I2 weeks
`(range l0-I4 weeks; visit 2) and 24 weeks (range 22-26
`weeks; visit 3) of MTX therapy. Patients were withdrawn
`from the study if they missed more than 3 weeks of MTX
`treatment for any reason.
`Clinical aasassrnent. Each patient was examined and
`interviewed by the same physicianlnutritionist (SLM) and
`rheurnatology research assistant (WA or PKY). Medication
`compliance was determined by direct questioning. and by
`pill counts when possible.
`'i'he following clinical variables (5) were determined
`at each visit:
`l) the number of joints with swelling (of 58
`diarthrodial joints); 2) the number ofjoints with tenderness
`on pressure. pain on passive motion. or both (ol'60joints): 3)
`the joint swelling index, expressed as a sum, in which each
`joint was graded for swelling as 0 (none).
`I
`(mild). 2
`(moderate). or 3 (severe): 4) the joint tendernesslpain index.
`expressed as a sum with joints also graded on the above
`scale ofo (none) to 3 (severe); 5) mean grip strength for both
`hands; 6) the duration of morning stiflhess. to the nearest
`hour: 7) the patient’: assessment of disease activity. graded
`as 0 (asymptomatic). 1 (mild). 2 (moderate). 3 (severe). or 4
`(very severe); 8)
`the physician's assessment or disease
`activity. graded on the same scale: 9) l-day dietary recall.
`
`Age
`Maleslfemales
`Previous use of folate-
`containing vitamins?
`Yearsotdisease
`Rheumatoid factor
`positive
`initial scrum folate. ng/ml
`
`initial RBC folate. nyrnl
`
`Conconent use of
`NSAIDsIaspirin
`Concurrent-use of
`prednisone
`Anatomic staacl
`Catpalzmetacarpal ratio!
`
`52.0 :2 14.6
`3Il3
`S
`
`8.7 -.t 5.5
`I!
`
`8.0 2 5.0
`(2.0-20.9)
`355.9 :2 222.8
`(101439)
`l4
`
`50.9 : I3.5
`3ll3
`2
`
`I53 1 ll.03
`I6
`
`8.5 :t: 6.0
`(L3-23.1)
`36l.3 : l93.4
`046-772)
`l6
`
`9
`
`B
`
`3.6 2 1.3
`0.50 : 0.03
`
`3.9 tr l.l
`0.51 t 0.03
`
`‘ Vshies are the mean : SD or the number of patients. Numbers in
`patenthescsare ranges. RBC - red blood cell; NSAlDs = nonster-
`oidal antiinflammatory rings.
`1 Mean 400 ug folatelday.
`3 P < 0.05.
`i For explanation. see refs. l4-I6.
`
`definite or classic disease (I3) gave informed consent and
`entered the study, which was approved by the Institutional
`Review Board of The University of Alabama at Birmingham.
`Seven patients could not be included in the data analysis: 4
`because of noncompliance with the MTX regimen. 2 because
`of administration of intramuscular MTX, and 1 because of
`self-medication with large amounts of FA. None of these 7
`patients were withdrawn because of the occurrence of toxic
`manifestations.
`hie and clinical characteristics of
`the 32 patients who completed the study are presented in
`Table 1.
`Criteria for entry into the study included RA oi‘ more
`thano months duration. with onset alter I6 years of age. and
`at least 3 of the following: 23 swollen joints. :6 or more
`tender joints, 245 minutes of morning stiflness. and a
`Westergren erythrocyte sedimentation rate (list!) of 228
`mmlhour. itadiosnph; were read by the project rheumatoi-
`ogist (GSA) in a blinded maruter. ‘file anatomic yading
`criteria of Bercns and Lin ( lot). as modllled by Trentham and
`Masi (is). as well as the carpa.l:metscarpal (C:MC) ratio (15).
`were used to assess the radiographs. These measurements
`had been previously used and validated at our institution and
`had been found to be reliable ()6).
`Patients with serious concomitant medical illnesses
`such as cancer. liver or renal disease. liver enzyme levels
`more than twice the upper limit of normal. white blood cell
`(WBC) counts <3.50tllmm’. or platelet counts <lS0.000I
`min’ were excluded from the trial. Previous use of MTX
`within the past 6 months and treatment with total lympho-
`cyte irradiation were also exelusion criteria. Patients were
`not accepted into the trial until sold salts. D-penicillamine.
`
`Kopie von subilo a.V.. gelieien far Maiwald Patentanwalts GmbH (COM06X00702)
`
`Lilly Ex. 2084
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2084
`Sandoz v. Lilly IPR2016-00318
`
`
`
`FOLIC ACID SUPPLEMENTS IN MTX-TREATED RA
`
`11
`
`using food models, which was coded using the Ohio State
`Nutrient Data Base (Unison Systems Associates, Colum-
`bus. OH); 10) current medications and dosaaea: and l 1) toxlc
`side effects. Before beginning MTX. each patient was coun-
`seled by his or her attending rheumatologlst with regard to
`possible toxic side effects. The presence of such side effects
`was investigated by asking the question. "Is the MTX
`medication causing you any problems?" Toxicity was also
`noted and recorded every 3 weeks by telephone interview.
`A toxicity score (T8) was determined for each pa-
`tient. The scoring system was based on the clinical experi-
`ences of the investigators and represents an attempt to
`distinguish mild and marginal toxic sympto (e.g.. slope-
`cia) from ‘severe and medically important ones (e.g.. cytope-
`nias) (10). As shown below, it was designed to increase in
`proportion to the duration of toxic events, their intensity.
`and their clinical significance, and to decrease with the time
`on the protocol at which the toxic manifestations first
`appeared (i.e.. ifs toxic reaction occurred early in the course
`of treatment. this would result in a higher toxicity score than
`if the same reaction occurred only after a longer course of
`treatment). The duration of the toxic event was placed in the
`numerator of the toxicity score, in order to quantitatively
`emphasize persistent morbidities while minimizing the con-
`tribution of transitory morbidities and spurious abnormal
`laboratory values. The T8 was calculated as follows:
`
`TS _ S
`
`(duration of toxic events [weelts]) x (intensity)
`
`X (clinical severity factor)
`weeltson protocol
`
`where intensity = 1 (mild). 2 (moderate). or 3 (severe): and
`clinical severity factor = I (alopecia. nausea. pruritus.
`anorexia andlor general 01 intolerance lpyrosis. cramps.
`ctc.]). 2 (vomiting. diarrhea, stomatitis andlor rash), 3 (ele-
`vated liver enzyme levels andlor elevated serum creatinine
`level), or 4 (cytopenia. documented infections. andlor pul-
`monary toxicity).
`Abnormal results on liver function tests were defined
`as transaminase andlor alkaline phosphatase values >2 times
`the baseline levels; cytopcnia was defined as a WBC count
`<35.000Imm’ or a platelet count <l50.000Imm’; elevated
`serum ereatinine level was defined as >15 mgldl; and
`pulmonary toxicity was defined as evidence of new intersti-
`tial pulmonary infiltrates compared with the baseline chest
`radiograph. with evidence of restrictive changes seen on
`pulmonary function tests (i.e.. vital capacity <80% of pre-
`dicted. normal expiratory flow rates. normal maximal vol-
`untary ventilation. and carbon monoxide difiusing capacity
`<‘I0% of normal [l1]). infection was defined as a docu-
`mented viral. bacterial. or fungal infection that compromised
`the patient's condition significantly. requiregd hospitaliza-
`tion. andlor required administration of systemic antibiotics
`or antifungal agents. Any abnormality in transltminase level.
`alkaline phosphatase level. WBC count. or platelet count.
`documented infection believed to be related to the MTX. or
`pulmonary toxicity was given an intensity score of 3
`(severe).
`.
`Over-all response to treatment was determined by a
`modification of the criteria developed by the Cooperative
`Systematic Studies of the Rheumatic Diseases group (18.
`
`‘lfahle 2. Mean : SD cumulative doses ofmethotrexate (MTX) and
`
`dietary intalte of folate and vitamin 8., - '
`
`Visit I
`Visit 2
`Visit 3
`
`Cumulative MTX dose. ms
`Folatcgroup
`Placebo group
`Dietary folate intake.
`udder
`Folate group
`Plaeebogtoup
`mm Vlllllifl Bu lull“.
`fiddly
`2.0 1: 18
`3.2 1 2.6
`2.! 2 L4
`l-‘olate group
`
`Placebo group 2.0 t l 2 2.2 t MI I 5 2 0 9
`
`
`
`-
`-
`
`8032] '
`70 2: 25
`
`I832“
`ISO 3: 37
`
`'
`
`I96 _+. H2
`206 t 176
`
`274 2 I89
`ISO .'t 90
`
`220 : I35
`ill! a: 53
`
`19). as follows. ‘l'he score on the joint swelling index and the
`joint tendemesslpain index at visit 2 andlor visit 3 was-
`compared with the score at visit I. Marked improvement
`was defined as a 250% decrease in these scores. moderate
`improvement was defined as a 314995 decrease in the index.
`no change was defined as a score remaining within 30% of
`the original value. and worsening was dellned as in >30%
`increase in the score. lmprovement and worsening in the
`physician or patient assessment of disease activity were
`defined as changes ofat least 2 integers on the 5-point scales.
`Laboratory assessment. At visit I. the complete blood
`cell count (CBC) including platelet count. Westergren ESR,
`liver enzyme levels (aspartate aminotransferase [AST] and
`alkaline phosphatase). RF titer. and serum creatinine level
`were determined. Followup CBC. creatinine studies. and
`liver lunetion tests were performed at the discretion of each
`patient’: rheumatologist and were generally repeated at each
`followup visit. Tests were performed more otten ‘(usually
`weekly) ifan abnormal value was obtained.
`lllochesnical assessment. At visits 1, 2, and 3. blood
`was obtained for a vitamin screen (serum and red blood cell
`[REC] folate. serum vitamin B”, vitamin C. vitamin A.
`fl-carotene. vitamin 8,. thiamine. and riboflavin) and for
`determination of the C. index. by methods previously de-
`scribed (20.2l). The C.
`index measures the activity of a
`folate-dependent enzyme system in peripheral blood motto-
`nuclear cells by assaying the formation of serine from
`glycine and radiolabeled formate.
`Statistical analysis. Either Student's I-test or analysis
`of variance followed by the least significant dilference test
`was used to compare means of normally distributed data.
`The Wilcoxon signed ranlt test was used with C, index data.
`Speannan‘s rank correlation test and chi~square analysis
`were used when appropriate (22). P values less than orequal
`to 0.05 were considered significant.
`
`RESULTS
`
`Characteristics of the patient groups. Sixteen of
`the 32 patients were in the folate supplement group.
`and I6 were in the placebo group. Findings from pill
`counts and questioning suggested a high degree of
`compliance with the MTX and FA/placebo regimen.
`
`Kopie von subiio e.V.. geliefert for Maiwald Palenlanwalts GmbH (COM06><00702)
`
`Lilly Ex. 2084
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2084
`Sandoz v. Lilly IPR2016-00318
`
`
`
`MORGAN ET AL
`
`Table 3. Patient response to methotrexate treatment‘
`
`Variable.
`patient mun
`
`Marked
`improvement
`
`Moderate
`improvement
`
`Marked or
`moderate
`improvement
`
`No
`change Worsening
`
`No. of patients
`
`-Swelling index
`Visit 2
`Folate
`Placebo
`Visit 3
`Palate
`Placebo
`Painltenderness Index
`Visit 2
`Folate
`Placebo
`Visit 3
`- Folate
`Placebo
`Joint swelling count
`Visit 2
`Folate
`Placebo
`‘ Visit 3
`Folate
`Placebo
`Joint painltenderness count
`Visit 2
`Folate
`Phceho
`Visit 3
`Polate
`Placebo
`Physician assessment
`Visit 2
`Polatc
`Placebo
`Visit 3
`Folate
`Placebo
`Patient essesunt
`Visit 2
`Pohte
`Placebo
`Visit 3
`Polate
`Placebo
`
`°'!‘heinitinlstudypopulati
`on eonaistedof 16 patients inthe rotate groupand I6 in the placebo group.
`Numbers shown tonl less than I6 for each visit because patients dropped out of the study. missed
`visits, or the variable was not determined.
`
`The mean duration of disease was longer in the placebo-
`treated group; other baseline demographic and clinical
`characteristics. including disease severity as measured
`byjoint counts. anatomic stage. and C:MC ratio. were
`not diflerent between the groups (Table I). There
`were no statistically significant diflerences between
`the groups in baseline hemoglobin or hematocrit
`values. mean corpuscular volume (MCV). WBC
`counts. or AST. alkaline phosphatase. or serum cre-
`
`atinine levels. The groups did not difier in their initial
`mean serum and RBC folate levels. In addition, there
`was no significant diference between the groups in
`their mean dietary intake offolatc or vitamin Bu. or in
`the cumulative MTX intake at any of the visits (Table
`2)» While there was a trend toward higher dietary
`folate consumption in the folate supplement group. it
`was not biologically important. given the much larger
`amount of FA that these patients received in the form
`
`Kopie von subilo e.V.. gellefen ltir Maiwatd Palentanwalts GmbH (COM06XO0702)
`
`Lilly Ex. 2084
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2084
`Sandoz v. Lilly IPR2016-00318
`
`
`
`FOLIC ACID SUPPLEMENTS IN MTX-TREATED‘ RA
`
`of the supplement. In addition, the patients‘ mean
`dietary intake of folate was below the recommended
`dietary allowance of 400 us/day (23).
`Eflleacy. Table 3 shows a comparison between
`the folate supplement and the placebo groups in their
`degree of response to MTX treatment. Chi-square
`analysis failed to show any statistically significant
`difi'erence between the 2 groups in the degree of
`improvement.
`Toxicity and patient dropout. Fifteen patients
`(41%) experienced some form of toxicity; of these. 10
`(67%) werein the placebo group, and 5 (33%) were in
`the folate supplement group. The toxic manifestations
`observed and the number of patients experiencing the
`reaction.- in the placebo group and the folate supple-
`ment group, respectively, were as follows: nausea 8
`and S. elevations in liver enzyme levels 3 and 0.
`cytopenia 1 and 1. anorexia 2 and 0. alopecia 0 and 2,
`constipationlbloating l and 0. pyrosis l and 0, stomach
`cramps 0 and I, stomatitis 0 and 1. and transiently
`elevated creatinine levels I and 0.
`
`
`
`TOXICITYSCORE
`
`noA
`
`000 00000 COO
`
`oofioo
`
`FOL ATE
`
`PLACEBO
`
`Figure 1. Toxicity scores in the folate supplement and placebo
`groups. Horizontal bars show the mean score: (0.2! and l.06.
`respectively: P = 0.027). sec Patients and Methods for explanation
`of toxicity score calculation.
`
`OCO
`
`-we0
`
`TOXICITVSal‘
`
`I00
`
`390
`
`I00
`4%
`llcfflflt
`
`I09
`
`700
`
`I09
`
`nuts 1. Correlation of toxicity score (TS; see Patients and Meth-
`ods) with initial plasma (A) and red blood cell (REC) (ll) folate levels
`(ndml) in the placebo group. Speannsn's ranlt correlation cosm-
`eient was -0.84 (P < 0.01) and -0.66 (P < 0.01) for plasma and
`RBC folste levels. respectively. Normal levels of plasma and R30
`folate are :2 up/ml and zldo nglrnl. respectively.
`
`Five of the patients (16%) dropped out of the
`study before visit 3. Four of these patients were in the
`placebo group; their toxicity scores were l.0, 2.67,
`3.0. and 5.1. MTX had to be discontinued in these
`patients because of toxicity. The toxic reactions ob-
`served in the 4 patients were severe nausea. anorexia.
`constipation, persistently elevated liver enzyme levels
`(2 or more determinations 1 week apart). cytopenia.
`and elevated creatinine levels. Elevated creatinine
`
`levels were attributed to MTX toxicity since they
`normalized following discontinuation of the drug. in
`the 1 patient in the folate supplement group who was
`withdrawn from the study. MTX treatment was
`stopped. not because of toxicity, but in preparation for
`surgical joint replacement.
`The mean toxicity score in the folate supple-
`ment group was significantly lower than that in the
`placebo group (Figure I). Minor toxicity (i.e., toxicity
`score <0.2) or no toxicity occurred in t2 patients in
`the folate supplement group, whereas only 7 patients
`in the placebo group were in this category.
`
`Kople von sublto o.V.. geliefen tor Maiwald Patentanwalts GmbH (COM06X0O702)
`
`Lilly Ex. 2084
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2084
`Sandoz v. Lilly IPR2016-00318
`
`
`
`MORGAN ET AL
`
`A significant increase over time in the mean MCV was
`not seen in the folate supplement group. The increase
`in the mean MCV was significantly higher in the
`placebo group than in the folate supplement group.
`Biochemical analyst. As seen in Table 4. there
`was a decrease over time in the mean C. index from
`visit I in the group of patients as a whole. but this was
`statistically significant only at visit 2. However, when
`the treatment groups were compared. the mean de-
`crease in the C, index in the placebo group was greater
`than that in the folate supplement group. Thus, the
`drop in the C. index reflects not only the expected
`MTX antagonism of folate-requiring pathways, but
`also the partial protection conferred by the simulta-
`neous administration of the FA supplement.
`A severe decrease in the mean C, index was
`associated with moderate toxicity, while minor or no
`toxicity was associated with no change in the C. index
`(Table 5). In patients with improved swelling and
`improved pain/tenderness indices (i.e.. marked im-
`provement or moderate improvement), toxicity was
`also associated with a severe drop in the mean C,
`index, while patients who showed clinical improve-
`ment without maior toxicity had only a moderate
`decrease in the C, index. Some degree of suppression
`of the C.
`index was associated with. and may be
`necessary for. etlicacy, regardless of toxicity. As
`shown in Table 5, patients with improved painl
`tenderness indices and patients with improved swell-
`ing indices had mean decreases in the C. index of 31%
`and 42%, respectively, whereas patients with no im-
`provement in these joint indices had little or no sup-
`pression of the C, index.
`
`DISCUSSION
`
`The efllcacy of low-dose methotrexate in the
`treatment of RA was not compromised by supplemen-
`
`Tsbls 4. Mean percent change in the C, index. from visit I’
`
`Group
`
`All patients
`Point:
`Placebo
`
`Visit 2
`
`-4] (26)?
`-18 (I2)
`-65 (M)?
`
`Visit 3
`
`- I2 (24)
`5 (I2)
`-34 (l2)$
`
`° ‘l'heC, index measures thesctivityotafolate-dependentenayme
`system in peripheral blood mononuelear cells (see refs. 20 and 21).
`Sisteenpatientslnthefolategroupand lfipatientsintheplaeebo
`groupwereevalustedstvisit t. Nun-lbersinpareathesesarenvalues
`stvisltszandll.
`tP<o.olversusvlsit I.
`tP< 0.05 versus visit I.
`
`PLh_CE§0
`FOLATE
`FlgIIre3.ChIna'esinmeancorpusetrlarvolunse(MCV)(iIfl)fI°||I
`visitltovisitllin thefolatesupplementantlplacebo youps.
`HOlimnlIlbansIIo\vIhen|nnchIlne8(l.9fl[nolsignifis:Int]inllIc
`folIlcgroupandS.8ll[P<0.tl5]intheplacebogmup).Themean
`ehangeintilcplsceboglotipwassignilieantlyhigherthantltatislhe
`folatep-oup(P<0.lll).Mean:SDMCVvnlucsiIIthe!olategIoup
`sndthe placebo group. respectively. were 86.4 1 8.Ulandu.o :2
`8.3lntvisitl.IlId89.3:t:8.Sflantl89.81-6.2llatvisit3.
`
`The toxicity scores in the placebo group
`showed significant negative correlation with the initial
`plasma (Figure 2A) and RBC (Figure 2B) folate levels.
`'lhese correlations were not found in the folate sup-
`plement group, even though 4 patients in this group
`had low-normal initial plasma folate levels (i.e., <3.5
`nglml). Of the 4 patients in the placebo group who
`discontinued MTX treatment. 3 had initial plasma
`folate levels of <3.5 ng/ml, while the remaining patient
`had a level of 5.3 ng/ml.
`Laboratory finding. The mean values for hem-
`atoerit, WBC count, platelet count. creatinine. AST.
`and alkaline phosphatase did not differ significantly
`between or within the groups at any of the visits.
`Compared with visit I, there was a significant increase
`in the MCV at visit 3 in the placebo group (Figure 3).
`
`Kople von suhito e.V.. geliefert ltlr Maiwald Patentanwalts GmbH (COM06X00702)
`
`Lilly Ex. 2084
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2084
`Sandoz v. Lilly IPR2016-00318
`
`
`
`FOLIC ACID SUPPLEMENTS IN MTX-TREATED RA
`
`l5
`
`Table 5. Mean percent change in the C. index from visit I. by
`degree of toxicity or improvement with rnethotrexate treatment‘
`
`Minor or no toxicity
`Moderate toxicity
`
`Swelling index inr'proved-
`Swelling index not improved
`
`Pain and tenderness index improved
`Pain and tenderness index not improved
`
`Swelling index lmprovedlminor or no toxicity
`Swelling index improvedlmoderate toxicity
`
`Pain and tenderness index improved/minor or no
`toxicity
`Pain and tenderness index improvedlmoderato
`toxicity
`
`0 (30)
`-54 (Nil
`
`-42 128)’!
`5 (22)
`
`-3! (25)\‘
`-23 (25)
`
`-24 (11)
`-60 (l I)!
`
`-16 (I7):
`
`-SI (8)?
`
`' The C, index measures the activity ofa folate-dependent enzyme
`system in peripheral blood mononuclesr cells (see refs. 20 and 2!).
`Values are the mean from visits 2 and 3 combined. Numbers in
`parenthaes are n values used to calculate the mean. Moderate
`toxicity was defined as a toxicity score >0.2 (see Patients and
`Methods). improvement injoint indices wu defined as a decrease of
`>30% from visit I.
`i P < 0.01 versus visit I.
`3 P < 0.05 versus visit I.
`
`tation with daily FA during the period investigated in
`this trial. i.e., the flrst 6 months of therapy. Since
`35-40% of the US population uses nonprescribed.
`over-the-counter vitamin supplements (24. 25). and
`FA is among the supplements that are available over
`the counter, this finding of a laclt of a detrimental
`chest is useful information for clinicians whose pa-
`tients are taking MTX. Our results with folic acid are
`in contrast with those of studies on the elfect of folinic
`
`acid (leucovorin) on MTX eflicacy (26-30). in I recent
`study. folinlc acid at 45 mglweelt (beginning 4 hours
`after MTX administration) lessened GI and other toxic
`manifestations. but completely negated the cflicacy of
`the MTX (26). Since folinic acid is a stable reduced
`folate (5-formyl-tetrahydrofolic acid) that bypasses
`dihydrofolate reductase inhibition, dosage levels and
`intervals of administration may be more critical than
`with the fully oxidized (i.e.. FA) form of the vitamin (30).
`Supplementation with PA.
`I mglday. signifi-
`cantly reduced toxic manifestations of treatment with
`low-dose MTX-. Although the placebo group had a
`longer mean duration of disease. the 2 study groups
`were comparable in all other parameters measured.
`suggesting that longer disease duration could not have
`accounted for the differences observed. The median
`weeltly dosage of MTX in the folate supplement group
`was 7.5 mg. Patients treated with substantially higher
`
`doses of MTX may require supplementation with >1
`mg of PA per day to reduce toxicity.
`There has not been a controlled trial of the
`
`efl'ect on MTX toxicity with the use of FA supplemen-
`tation at higher doses. Some investigators have pre-
`scribed FA at >1 mglday for M'l'X-treated RA patients
`(3l.32), but its use has been controversial because of
`potential elfects on disease response (33,34).
`The concurrent administration of a drug and its
`antagonist is not without precedent in RA. as indicated
`by reports of successful D-penicillarnine therapy with
`vitamin B, supplementation (35,36). It should be
`pointed out that PA is a weaker MTX antagonist than
`is leucovorin. Therefore, concerns that FA supple-
`mentation will cause loss of therapeutic eflicacy of
`MTX are not well founded.
`
`Low-nonnal initial folate levels in our placebo
`group were correlated with a high probability of even-
`tual MTX toxicity, strongly suggesting that MTX
`therapy in a folate-depleted patient predisposes that
`individual to toxicity. A similar correlation between
`initial folate status and MTX toxicity has been ob-
`served in patients treated for tumors of the head and
`neck (37). Such a conelation was not found in our
`folate supplement group. because patients with mar-
`ginal baseline folate levels in this group underwent
`repletion of their folate stores concurrently with MTX
`therapy. Therefore, low-normal folate status should be
`included among ‘other well-recognized factors (38-42)
`as a contraindication for MIX treatment in patients
`with RA. This observation tal't_'es on "added importance
`since marginal folate nutriture may be fairly common
`in RA (43,44). it is also likely that marginal folate
`nutriture is an unrecognized determinant of toxicity in.
`and hence a contraindication for the initiation of MTX
`therapy for. psdiiasis-(45), asthma (46), and a variety
`of other conditions (4.7).
`Analysis of larger numbers of patients over time
`will help io determine if M‘!‘X‘toxicity is less likely in
`patients with normal éinitial folate status; it is known
`that folatge staltits.-becomes iarpaired with long-term
`administration of M'l'=x (20). The impaired utilization
`of folatcs in patients with vitamin 8., deficiency (48)
`would suggest that vitamin B5, leveis should also be
`checked :prior to initiation of MTX theravlt. although
`the importance of low 3,, values in RA is not com.
`plctely irriderstood (49).
`The data in ‘fable 4 suggest that careful adjust-
`ment of the extent of folateantagonism and of the
`dosageof FA is necessary in order to maintain clinical
`
`Kopie von subilo e.V.. gelieten ltir Malwald Patentanwalts GmbH (COM06X00702)
`
`Lilly Ex. 2084
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2084
`Sandoz v. Lilly IPR2016-00318
`
`
`
`16
`
`MORGAN ET AL
`
`response without toxicity. Patient improvement. as
`judged by joint indices, was associated with a decrease
`in the folate-dependent enzyme activities reflected in
`the'C, index. whereas those patients with no improve-
`ment had a smaller decrease. or even an increase. in
`the C, index. jrhis is consistent with the fact that
`sulfasaiazine. also an inhibitor of folate-dependent
`enzymes including those involved in the C. index, is.
`like MTX, useful in the treatment of RA (50-52). The
`association of severe folate antagonism with toxicity is
`demonstrated by the fact that the C. index dropped to
`significantly lower levels in the patients with toxicity,
`versus those with no toxicity. Clinically. this is con-'
`sistent with reports of cytopenia in patients receiving
`both trimethoprimlsulfarnethoxazole and MTX for
`RA._since this combination of 2 antifolates would be
`expected to be particularly toxic (53-56). A moderate
`drop in the C. index is associated with improvement in
`joint involvement, with minor or no toxicity.
`The C. index is an appropriate measure of the
`efl'ect of MTX on folate metabolism in RA. since
`peripheral blood mononuclear cells are immunologi-
`cally active and likely to be metabolically related to
`the target tissue in this disease. Furthermore. in these
`cells. the only source of folate-bound. I-carbon frag-
`ments (required for the synthesis of nucleotides) is the
`p carbon of serine. Since peripheral blood mononu-
`clear cells have a very limited capacity to synthesize
`serine from glucose. they may be particularly sensitive
`to MT}! inhibition of folate-mediated serine metabo-
`lism ($7).
`The precise mechanisms by which low-dose
`MTX produces therapeutic response andlor toxicity
`are unknown. Antiintiammatory and antiproliferative
`efiects have been postulated (58, 59). Some depletion
`of folate-bound l-carbon units and inhibition of folate-
`
`dependent enzymes may contribute to etllcacy by
`altering inflammation and proliferation in arthritis tar-
`get tissues (60.61). Excessive inhibition and depletion
`may cause toxicity in non-target tissues such as the
`liver and the GI tract (62.63). The greater increase in
`MCV in our placebo-treated group reilects an emerg-
`ing megaloblastic condition in the bone marrow, from
`impaired DNA synthesis (64).
`We believe that folic acid su