Medical Hypotheses 81 (2013) 942–947

Contents lists available at ScienceDirect

Medical Hypotheses
journal homepage: www.elsevier.com/locate/mehy

Challenging the clinical relevance of folinic acid over rescue after high
dose methotrexate (HDMTX) q
Ian J. Cohen ⇑
Department of Pediatric Hematology Oncology, Schneider Children’s Medical Center of Israel and Sackler Faculty of Medicine, Tel Aviv University, Ramat Aviv, Israel

a r t i c l e i n f o a b s t r a c t

Article history: The hypothesis: The use of adequate folinic acid rescue (in clinically relevant doses) after high dose meth-
Received 16 May 2013 otrexate will prevent neurotoxicity without reducing treatment results.
Accepted 25 August 2013 Methods: A literature search was performed to test the hypothesis that no evidence for the existence of
folinic acid over-rescue of high-dose methotrexate (MTX) in clinically relevant situations exists (evidence
that too much folinic acid reduced cure rate).
Empirical data: Examples of folinic acid over-rescue after lower doses of MTX were found and has been
cited as evidence of over rescue of high dose MTX. Mega doses of folinic acid, used when toxic levels of
MTX occurred, also could neutralize the MTX effect. Data were found to support the contention that
higher levels of MTX require disproportionally higher folinic acid doses for rescue. Careful examination
of the available studies after HDMTX yielded more convincing alternative explanations for reduction in
cure rate than over rescue. Little convincing evidence for the existence of over rescue after HDMTX
was found.
Discussion: The rescue of high-dose MTX with an appropriate dose of folinic acid that can prevent toxic-
ity, especially neurotoxicity, was not shown to reduce the therapeutic effect. No evidence was found that
higher doses of folinic acid after high dose MTX reduces the therapeutic effect.
Consequences of the hypothesis: Acceptance of the hypothesis can prevent harm being caused (especially
brain damage) by reversing the trend of dose reduction in FA rescue. The recognition that the use of
higher folinic acid doses is safe, can prevent neurotoxicity, and does not reduce prognosis has important
implications for the development of effective non toxic treatment protocols.
Ó 2013 Elsevier Ltd. All rights reserved.

Introduction
High dose MTX (HDMTX) shows continued effectiveness in the
treatment of childhood acute lymphatic leukemia (ALL) [1], osteo-
sarcoma [2] and brain tumors [3]. Higher doses are often more
effective than lower doses [4]. Neurotoxicity seems related to the
dose of folinic acid (FA) rescue used [5]. The use of ‘‘too large’’ res-
cue doses of folinic acid (leucovorin (LCV), LV, citrovorum factor)
after HDMTX has been blamed for rescuing malignant cells and
negating the MTX chemotherapy effect [6]. The belief that over-
rescue of high dose MTX (HDMTX) exists, in clinically relevant
doses, has resulted in progressively lower FA rescue doses being
used. This ignores the danger of increased neurotoxicity, resulting
in a growing incidence of MTX neurotoxicity in childhood malig-
nancies, especially (ALL) [7]. This study was performed to see if
the published data supports the hypothesis that, in clinically rele-
vant doses, over rescue of HDMTX by FA does not exist.
The meaning of some terms have changed with time. HDMTX
was defined as >1 g/m2 [8], <500 mg/m2 low dose MTX, and 0.5–
1 g/m2 intermediate dose MTX. Today the term HDMTX means at
least 5 g/m2, medium dose MTX 0.5–5 g/m2 and low dose
MTX < 500 mg/m2. It is also useful to define a dose of FA. Over
1 g/m2 as ‘‘Mega Dose FA’’ (used after renal shutdown), over
315 mg/m2 as high dose FA, (considered ‘‘too much’’ by Borsi [6])
and between 45 and 315 mg/m2 as contemporary FA rescue doses
without regard to adequacy.
Many original studies were performed over 40 years ago, and
contemporary researchers may only have abstracts or citations
from other manuscripts both of which may be misleading.

The hypothesis
q
No funding was given in support of this manuscript.
⇑ Address: Department of Pediatric Hematology Oncology, Schneider Children’s The use of adequate folinic acid rescue (in clinically relevant
Medical Center of Israel, 14 Kaplan St., Petach Tikva 49202, Israel. Tel.: +972 3 925 doses) after high dose.
3781; fax: +972 3 925 3042. Methotrexate will prevent neurotoxicity without reducing
E-mail address: icohen@tau.ac.il treatment results.

0306-9877/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.mehy.2013.08.027
 I.J. Cohen / Medical Hypotheses 81 (2013) 942–947
Evaluation of the hypothesis
A detailed electronic search was performed of articles indexed
by Pub Med Advanced (any date) and Google Scholar (from 1992
to 2012) using the index terms ‘‘over-rescue,’’ ‘‘leucovorin’’ and
‘‘folinic acid’’ relevant articles were obtained and reviewed. All
articles cited as supporting over rescue were examined. In addi-
tion, relevant articles were examined from a personal database
of more than 460 article reprints from 132 journals on metho-
trexate, folinic acid, and side effects. Articles found suggesting
that higher doses of FA reduced prognosis were reviewed, and
articles they cited to support the claim noted, and reviewed.
Empirical data found and analyzed
To understand the significance of the findings reported here, it
is important to review several significant reports that are often ci-
ted to support the concept of over rescue.
Sirotnak et al. [9] mention that ‘‘a progressive increase in the
calcium leucovorin dosage on any schedule reduced both the tox-
icity and the antitumor effect of MTX’’. However the manuscript
also reported that ‘‘On certain schedules with MTX toxicity could
be virtually eliminated with no diminution in anti-tumor effi-
cacy.’’ They found that in mice inoculated with L1210 leukemia
cells or S180 sarcoma cells, treatment with MTX 400 mg/kg (about
12 g/m2) followed by FA 120 mg/kg (leukemia) or 72 mg/kg (sar-
coma) increased the life span by 87–94% (leukemia) and 81–
113% (sarcoma), with a minimum of toxicity-related deaths (0/
30 and 1/30, respectively). This article also shows that the sim-
plistic concept of a linear FA/MTX ratio for adequate rescue, ac-
cepted by several authors [10–13], is not valid. When the dose
of MTX was doubled, the dose of FA needed to maintain the in-
creased life span and avoid toxic deaths rose by 3.3-fold in the
leukemia group (to 400 mg/kg) and 4-fold in the sarcoma group
(to 288 mg/kg). Pinedo [14] and Boarman [15] Koizumi [16] re-
ported similar findings. Bertino [11], noted that, ‘‘If LV administra-
tion is delayed for more than 42 or 48 h following HD MTX, severe
and irreversible toxicity may occur’’. Bleyer [17] stated that ‘‘If
HDM infusions are continued for more than 42 h before citrovo-
rum factor is initiated significant myelosupression and gastroin-
testinal mucositis will occur.
The empirical data has been analyzed in several sub groups. All
articles cited in support of the concept of over rescue have been
noted and they have been added to the articles analyzed even if
they are reports of over rescue following medium or low dose MTX.
Over rescue by mega dose FA after HDMTX
A case report [18], reported mega dose FA ‘‘over-rescue’’ after
high-dose MTX. A patient with osteosarcoma received an inappro-
priately high dose of FA (1275 mg) after her second treatment
with 27.5 Gm of MTX (15 g/m2) owing to a misunderstanding
about the timing of the 6-h serum MTX level. The tumor became
painful and started to grow again. The only clinical situation in
which such a dose would deliberately be used is after renal shut-
down following high-dose MTX (with prolonged toxic levels).
Nevertheless, clinicians’ fears of causing death by toxicity and
the subsequent legal and financial implications might push aside
all worries that mega dose over-rescue can cause relapse and
death [19].
Sirotnak [20] showed in a mouse model that following 400 mg/
kg MTX (equivalent to 12–15 g/m2), rescue at 16 h with 24 mg/kg
FA (equivalent to 720 mg/m2) gave a 2 log better cell kill than after
rescue with 400 mg/kg (equivalent to 12,000 mg/m2) FA.
943
Over rescue of folinic acid after high dose MTX
Borsi et al. [6] published a study entitled, ‘‘How Much Is Too
Much? Folinic acid rescue doses in children with acute lymphatic
leukemia.’’ Seventy-one children aged 1–18 years with ALL, were
treated with 6–8 g/m2 MTX followed 36 h later by 75 mg FA and
then at least another 16 doses of 15 mg FA. Recalculation, by body
surface area (range 0.3–2.0 m2) revealed that the patients had re-
ceived a total of 157.5–1050 mg/m2 FA. This data is not included
in the abstract. There was a non-statistically significant trend for
better prognosis with less than 315 mg/m2 FA. However, given that
this dose far exceeds (by about 700%) the 45 or 90 mg/m2 total
dose of FA currently administered after 6–8 g/m2 of MTX in pediat-
ric ALL, why this article is cited to support the current doses re-
mains obscure. Indeed, several years later, Borsi’s group reported
that, they had reduced the FA to approximately 230 mg/m2 after
6–8 g/m2 of MTX [21].
Skarby et al. [23] expressed concern that ‘‘high leucovorin doses
during HDMTX treatment may reduce the cure rate in childhood
ALL,’’ careful analysis points to a problem in blaming the FA dose
for the increased relapse rate. Although the abstract states that
for all the groups together, a higher FA dose increased the relapse
rate, the article itself states ‘‘the median leucovorin dose of all
courses was higher in patients with relapse in the IR [intermediate
risk] and HR [high risk] groups as opposed to the SR [standard risk]
and VHR [very high risk] groups’’. Thus in two of the four risk
groups, a higher dose of FA was associated with less relapses than
a lower dose. The number of patients in the IR and HR groups was
244 and in the SR and VHR groups 201.
Graff et al. [25] describes treatment of 19 osteosarcoma patients
who received 130 courses of MTX 12 g/m2 (max 20 g/m2) with FA
rescue. Patients achieving a mean serum level at the end of the
infusion of 1300 micromoles of MTX did better. In relapsing pa-
tients 27% (7 of 26) of the therapy courses included increased doses
of FA compared with 9% (9 of 103) who did not. The most signifi-
cant factor for relapse was increased FA rescue on univariate and
multivariate analysis. This demonstrates the difficulty in analyzing
such data since each patient received several treatments and there
was a difference between the levels reached each time. The level of
MTX on each occasion was tested for correlation with prognosis
but the contribution of each treatment is unclear. A higher mean
serum level correlated with better prognosis. Four patients re-
lapsed, three of whom had mean peak serum MTX less than 1000
micromoles. A most unexpected finding was that they were the
group who received increased doses of folinic acid. One explana-
tion could be that patients with the lowest levels of MTX at the
end of the infusion had the most prolonged high MTX levels .This
seems most unlikely since the opposite was found by Crews
et al. [26] who noted that increased doses of folinic acid were
found in the patients with the highest MTX levels at the end of
the infusion. One suspects that most or all of the seven treatment
with increased folic acid rescue given to the relapsed group were
given to the patient with a very high MTX peak level, putting in
doubt the significance of the statistics for the whole group.
Crews et al. [26] found patients with Osteosarcoma with the
highest MTX levels had a worse EFS, explainable by delay of the
next treatment or a reduction in dose intensity. Since patients with
the highest MTX levels received more folinic acid ‘‘another possible
explanation is that increased leucovorin dosing in patients with
very high MTX exposures may have compromised the antitumor
effect of MTX’’. This explanation is not necessary since delay in
treatment and reduced dose intensity has been shown to reduce
prognosis in other studies [27].
Wolfrom et al. [24] state that after HDMTX (defined as 1–12 g/
m2) ‘‘the intensive LCV rescue needed to avoid serious toxicity may
also be beneficial to malignant cells’’ No data is presented to
944 I.J. Cohen / Medical Hypotheses 81 (2013) 942–947
support this conjecture. ‘‘The high dose of LCV used in HDMTX re-
gimes may be advantageous for lymphoblasts, thus LCV rescue
should be as low dose as feasible and initiated as late as possible’’.
Again no data is given to support this conjecture. They add the
comment ‘‘whether a reduced and delayed LCV rescue would im-
prove efficacy of HDM infusions has to be proven by further stud-
ies.’’ In the article two protocols for first relapse in Childhood ALL
were compared that included in induction, 1 g/m2 MTX over 36 h
followed by 15 mg/m2 FA after 48 and 54 h from start (total
30 mg FA) or 12 g/m2 MTX over 4 h and 15 mg/m2 FA q 6 h  12
starting 24 h from the start of MTX (total 180 mg/m2).The first
group had deep mucosal ulcerations in 41% and a median treat-
ment delay of 3 days, the other group did not. The results were
similar in both groups 95% achieved remission after completion
of induction. Since the results were similar in both groups the only
reasonable conclusion is that intermediate dose MTX with minimal
rescue is as good as HD MTX with rescue but more toxic.
Janka [27] in a report of patients with high risk ALL in the
COALL-82 trial showed no benefit from the addition of high dose
MTX to early intensification. The patients had received three
courses of 100 mg/kg MTX over 4 h followed at hour 24 by FA
0.5 mg/kg 10 q 6 h (personal communication G E Janka).This is
roughly equivalent to 3 g/m2 MTX followed by 150 mg FA. They
were worried ‘‘whether the ample dose of folate may have had a
detrimental influence on the prognosis’’. Although this is an under-
standable worry the other changes in chemotherapy doses and
schedules, from the previous COALL-80 protocol (reduction in
Anthracycline and Cyclophosphamide doses and change in Aspara-
ginase schedule) prevent any conclusions being drawn.
Hegyi [2] in an article on MTX pharmacokinetics in Osteosar-
coma mentions over rescue, citing Crew [26] and emphasizes the
risk of leucovorin over-rescue but provides no supporting data.
The POG 9005 study
Mahoney [4], as part of the POG 9005 trial, compared children
with B-precursor ALL. Groups A and C received IV MTX 1000 mg/
m2 for 24 h followed by FA 5 mg/m2 every 6 h  5, starting 48 h
after the start of MTX. Group B received oral MTX 30 mg/m2 every
6 h  6 (180 mg/m2) and FA 5 mg/m2  5 starting 48 h after start-
ing MTX. Severe neurotoxicity occurred in 8.0% in group A (of
whom 57.1% had leukoencephalopathy) and 3.7% in group B (of
whom 15.4% had leukoencephalopathy). Reviewing this study,
Mantadakis [28] suggested that the patients who received the low-
er oral dose of MTX were over-rescued because their results were
worse, although they had less mucositis and hematologic toxicity.
Although Mahoney. [12] found the reason for the increased neuro-
toxicity with regimen A to be unclear, applying the data of Bertino
[11] reported above, the rescue by 5 mg/m2 FA at 48 h and 20 mg/
m2 thereafter to be too little too late, resulting in similar neurotox-
icity to that seen without any FA rescue at all. Mahoney reported
[4], that regimen B was less effective, though with lesser neurotox-
icity than in groups A and C. Because the MTX dose was lower. The
resulting neurotoxicity resembled that seen after 180 mg/m2 MTX
without FA rescue. Thus, Mantadakis’s explanation did not take
into account the expected worse results and lower toxicity of the
lower (virtually non-rescued) dose of MTX relative to the higher
(also non-rescued) dose. In a subsequent report, Mahoney [29] de-
tailed the side effects of treatment and included the finding that
group C (not unexpectedly) had a high rate of serious neurotoxicity
(8.1%), similar to group A. He was concerned that the 8/168 re-
lapses in the IV MTX group (4.7%) as opposed to 9/178 relapses
(10%) in the low-dose oral MTX group may have affected the over-
all complete remission rate. He argued that if the 10 mg/m2 dose of
FA in the study by Winick et al. [31] was sufficient to rescue MTX
100 mg/m2, then the dose of 25 mg/m2 FA administered after
180 mg/m2 MTX in the study of Mahoney et al. [4] was too high.
However, the FA was started 48 h after MTX, and so would have
had little rescue effect. Thus, the lower response rate was probably
a result of the treatment and not a confounding factor, and the dif-
ference in the FA/MTX ratio had little relevance. The data in the ori-
ginal study [4] and others cited [31–33] (discussed below) do not
support the claim that ‘‘merely reducing the leucovorin in the PO
MTX group might have made this arm as effective as the IV MTX
arm’’ [30]. Mahoney notes that for LV rescue ‘‘there is no estab-
lished ratio that define safety and efficiency. Under rescue with
LV may pose a different threat to children who receive intensive
MTX schedules’’ [34].
Salzer [35] examined plasma and red blood cells MTX levels in
POG 9005 and 9006 patients remarking that ‘‘specifically, excessive
delay or the use of ‘‘low doses of leukovorin’’ were associated with
toxic deaths whereas large doses of leukovorin or the delivery of
rescue shortly after the completion of the MTX therapy interfered
with efficiency.’’
Over rescue after medium MTX
Joannon [36] reported the ALL PINTA 1992 protocol that used a
24 h infusion of 1 or 2 g/m2 MTX in medium risk patients followed
by 15 mg/m2 FA at 42, 48 and 54 h from start of the MTX. The FA
dose was according to the creatinine clearance and was started
at 42 h after 2 g/m2 MTX. When the creatinine clearance was equal
or more than 100 ml/min rescue was started at 48 h. (total dose of
FA was 30 mg/m2). The doses of FA given when rescue was started
after 42 h ranged from 45 to150 mg/m2.They were worried that
over rescue reduces the antitumor effect of MTX. that They felt
‘‘the leucovorin rescue was excessive, compared to the BFM proto-
col, in a substantial portion of the MTX doses. Perhaps this might
be one of the factors that contributed to the lower event free sur-
vival (EFS) of the ALL PINDA 92 trial.’’ One can be forgiven for
thinking that the lower dose of MTX (1–2 g/m2 instead of 5 g/m2)
is a more convincing reason.
Jurgens. [37] reported that patients who received <50 mg/m2 cf.
>60 mg/m2 FA after a 24 h infusion of 500 mg/m2 MTX (together
with IT MTX) had a non significant trend toward better survival.
There was however a difference in outcome (p < 0.05) between
the 31 patients who started FA rescue before hour 43, 42% of whom
had an EFS at 8 years and the 45 patients who started rescue after
hour 43 (median 48 h) with an EFS of 76%. Literature cited, sup-
ports the importance, not of dose, but of time to starting FA.
Over rescue after low dose MTX cited to support over rescue after High
dose MTX
Winick reported the results of a protocol [31] details reported
elsewhere [38]. It contained 28 courses of oral MTX 25 mg/m2 q
6 h  4 followed 2 h later by IT triple therapy (MTX, Ara C and
Hydrocortisone) and 24 h later, 5 mg/m2 FA. Neurotoxicity oc-
curred and continued even when the IT triple was changed to IT
MTX. Then FA 5 mg/m2 q 12 h  2 was added 24 h after the last
dose of MTX and after IT MTX and prevented all neurotoxicity.
They were worried that rescue after 24 h from start of IT MTX
rather than after 48 h may have resulted in the relatively high
CNS relapse rate. A more reasonable explanation would be that re-
peated low dose MTX is not as effective as High dose MTX [40].
Browman. [22], noted that patients with head and neck cancer
who received 40 mg/m2 of FA after 40 mg/m2 of MTX had a lower
response rate than patients who received no FA and had marked
toxicity and more therapy delays.
Tishler [41] described over-rescue by FA after 7.5–12.5 mg
weekly MTX in patients with rheumatoid arthritis. Hills [42] and
Cipiano [43] showed a similar problem in psoriasis after 10–25
 I.J. Cohen / Medical Hypotheses 81 (2013) 942–947
MTX, and 50 mg MTX respectively. Takacs and Rodriguez [44] re-
ported that the failure rate of a single MTX dose in aborting ectopic
pregnancies was associated with a high folic acid serum level. The
dose of MTX, not mentioned in the article, was 50 mg/m2, (per-
sonal communication).
Sterba [8] looking at the pretreatment serum folate levels effect
on MTX serum levels subtitled the article ‘‘folate overrescue con-
cept revisited.’’ They do not suggest that there was any effect on
prognosis. The article by Valik [45] linking severe neurotoxicity
after high dose MTX with the pretreatment folate level also men-
tions over rescue.
CSF levels of FA following systemic FA
Opinions are divided about the effect of FA doses on CSF levels.
Strangely careful examination of the data shows a remarkable
agreement by all as to the levels actually achieved. Winick [38]
claimed that ‘‘the administration of oral leucovorin raises the con-
centrations of folate in the CSF and may selectively rescue cells in
the CSF.’’ On the basis of the data of Kamen and Vietti [33] showing
that CSF levels of FA reach 1.5  10 7 M, Winick [38] suggested
that ‘‘repetitive dosing of FA, such as given in standard intermedi-
ate and high dose MTX regimes, may significantly raise the CSF fo-
late for a prolonged time Therefore, the use of repeated oral FA to
rescue patients from the systemic side effects of methotrexate
should be of concern in the design of treatment regimes in which
anti-fols are used to treat or prevent malignant disease in the
CNS.’’ However, Kamen and Vietti [33] examined CSF folate levels
with a 5-methyl THF assay only at 6 h after a single exposure with-
out repeated dosing. By contrast, Allen et al. [46], using a different
microbiological assay, claimed that oral leucovorin cannot elevate
CSF 5-methyl-THF levels above 5  10 8 M, which was the level
achieved by Thyss. [39]. Further evidence of only a temporary in-
crease in CSF folate comes from a report by Mehta. [47], who also
used an atypical assay to show that 50 mg IV FA yielded a CSF level
of 2  10 8 M, and after a further dose 24 h later, the level rose
temporarily to 2  10 7 M and then rapidly dropped back. Thus,
whatever the opinions expressed, the MTX CSF levels in all these
studies were never above 2  10 7 M for more than a short time
and then rapidly returned to a level of 2  10 8 M.
Reviews commentaries and editorials
Mantadakis [48] expressed a worry that the concurrent use of
leucovorin may reverse the antileukemic effects of methotrexate
in the CNS, since folates are concentrated better than methotrexate
in the CSF. The article they cite [34] has been discussed above. They
felt that there was indirect evidence that too much FA after HD
MTX had a negative effect on clinical outcome in patients with
ALL by ‘‘comparison of the BFM86 [49] and AIEP-ALL 88 [50] stud-
ies who had very similar chemotherapy regimens and identical
definitions of leukemia risk.’’ Close examination of the protocols
however show other differences apart from leucovorin dose that
may well have effected treatment results especially the reduction
of two doses of doxorubicin and 7 days of dexamethasone in the
reinduction phase of the standard risk arm of the AIEP-ALL study.
‘‘the particularly poor results of the French Fralle 87 ALL Protocol
(10–15% lower event free survival than other synchronous studies
of children with ALL) were attributed by the investigators of the
study to the very high dose of leucovorin rescue of 150 mg/m2.’’
The article itself is less dogmatic and only states that this ‘‘may
be partially due to the very high dose of FA (150 mg/m2) used.’’
However they felt ‘‘the obvious explanation is the fact that there
was no intensification-like phase now part of the standard treat-
ment of ALL [51].’’ The contribution of the high doses of FA to
the poor prognosis is also questioned by the fact that similar doses
945
(180 mg/m2) of FA given after 3 and 8 g/m2 MTX (as in the Fralle 87
protocol) produced excellent results in the LMB89 protocol in B cell
lymphomas and L3 leukemia[52]. The review also notes the study
by Niemeyer et al. who reported better results after 4 g/m2 MTX
with 200 mg/m2 FA then 12 mg/m2 FA q 3 h compared with those
who received 40 mg/m2 MTX without rescue [40]. The same review
felt that patients who received 30 mg/m2 FA after 180 mg MTX in
the POG 9005 study were over-rescued compared with those
who received 1 g/m2 MTX with the same FA rescue. The evidence
for over rescue in the lower MTX dose group being the higher fre-
quency of mucositis and hematological toxicity in the other arm!
The neurotoxicity seen would suggest that the patients who re-
ceived 1 g/m2 MTX were more under-rescued than those who re-
ceived 180 mg/m2 MTX (see section on POG 9005).
Pui [53] analyzed the results of the CCG-191P study in which
randomized patients received very-high-dose MTX (33.6 g/m2
MTX) or 1800–2400 cGy cranial irradiation (plus intrathecal
MTX). Overall and event-free survival were similar in both groups.
The high-dose MTX group had significantly better neurocognitive
function over time (with increasing IQ scores over 10 years) but a
high cumulative risk of isolated central nervous system (CNS) re-
lapse. The authors speculated that the FA rescue might have been
given too early or at too high a dose, or that ‘‘the CNS-directed
therapy was inadequate’’. This last explanation is more convincing
since the first two explanations are based on the observation that
there was a lack of impaired neurocognitive function and the rela-
tively few systemic side effects in the very-high-dose MTX group,
reflecting the (baseless) fear that the FA dose needed to prevent
neurotoxicity will inevitably result in reduced efficacy.
Bertino in an editorial [32], ‘‘Leucovorin Rescue Revisited’’,
mentions that ‘‘too early rescue or too large a leucovorin dose
may decrease therapeutic gain’’ [9]. The report by Goldin et al.
[54] is cited to support this claim, However, Goldin et al. [54] com-
pared toxicity between FA administered simultaneously with MTX
or 12 h later. Different doses of MTX were used, but without men-
tion of the FA dose.
Vezmar et al. [55], in an analysis of the biochemical and clinical
aspects of methotrexate neurotoxicity states that, ‘‘Significant
neurotoxicity is observed in most high-dose chemotherapy proto-
cols with MTX including leucovorin rescue, suggesting that the
usual doses of leucovorin are not sufficient. . .’’ Nevertheless, the
same fear of folinic acid over-rescue is still apparent in their state-
ment, ‘‘. . .and a more intensified leucovorin rescue does not seem
to be judicious. . .’’ In another manuscript [56] The authors make
a less convincing suggestion that since the folate dose was ade-
quate in more than 90% of patients and they did not develop clin-
ical signs of neurotoxicity, ‘‘instead of increasing the rescue dose
for all patients, it might be a more rational strategy to identify
those patients who might benefit from more foliate.’’
Sterba [57] in a review discusses over rescue. Both extremes too
little and too much LV rescue could be harmful. The lack of
improvement in the UK ALLXI study when High dose MTX was
added is noted [58]. ‘‘One may speculate that late availability of
MTX results’’ may not have allowed down adjusting of ‘‘folinic acid
doses which, may have resulted in LV over rescue in many patients,
minimizing the potential systemic benefit of MTX but not its im-
pact on occult CNS disease as suggested earlier by Borsi [6]’’. Since
Borsi does not mention a differential effect on the CNS and sys-
temic disease, this does little to explain the 4% reduction in 5 year
CNS relapse with a 2% increase in non-CNS relapses [57].
Kamen and Weitman [59] comments on the article by Seidel
[21]who had used 6–8 g/m2 MTX then 50 mg/m2 (initially 75 mg/
m2) FA followed by 15 mg/m2 q 3 h  8 and then q 6 h until the
serum MTX level was below 5  10 8 (approximately 230 mg/
m2). They noted that ‘‘the lack of neurotoxicity, including the ab-
sence of even radiological findings in children with acute leukemia,
946 I.J. Cohen / Medical Hypotheses 81 (2013) 942–947
is pleasantly surprising, perhaps the lack of toxicity is secondary to
the extensive use of leucovorin rescue.’’
Kamen [60] remarks that early models and clinical experience
show that too much LV can reverse the anti tumor effect of MTX.
He cites another article by Sirotnak [20]. This article showed that
Leukemia 1210 cells were rescued 16 h after 400 mg/kg MTX only
with equimolecular mega dose FA rescue, 400 mg/kg FA(approx
8000 mg/m2) but intestinal cells could be rescued by 24 mg/kg FA.
Consequences of the hypothesis
This literature search yielded no convincing evidence of FA
over-rescue in doses currently used after high-dose MTX,
although it may clearly occur after low-dose MTX and after
mega dose FA rescue. Neurotoxicity correlates inversely with
FA dose. Lowering the dose of FA in patients receiving high-dose
MTX was not shown to improve the cure rate and is therefore
unjustified [61]. Recent reports have shown not only the efficacy
of high dose MTX in ALL [1] but also that the subtle neurotox-
icity that occurs after high dose MTX (12–20g/m2) in Osteosar-
coma patients can be prevented by higher doses of FA [5]. All
11 of the 18 neuropsychological assessments showing a differ-
ence between groups who received different FA doses favored
the patients who received more FA. Both FA dose groups, 300–
600 and 120–250 mg/m2, received higher doses than those used
in current ALL protocols [5]. The BFM and COG groups continue
to reduce the dose of FA rescue [37]. The NOPHO 2008 protocol
for all ALL risk groups stipulates that 42 h from the start of 5 g/
m2 MTX, only 15 mg/m2 FA should be administered, Given the
minimum of two doses, some patients will receive a total of only
30 mg/m2 FA (personal communication Arja Harila-Sari). Skarby
[23] states ‘‘the present results have had impact on the design
of the new protocol for treatment of ALL in the Nordic countries
(NOPHO ALL 2000) with a lowering of the LV dose as compared
to the previous protocol.’’ A literature analysis showed that the
doses of FA that prevent neurotoxicity would seem to be at least
180 after 6–8 g/m2 MTX [7]. The only study identified in which
different doses of FA were used without other concomitant pro-
tocol changes was reported by Winick [38]. A subset of 24 chil-
dren received increased FA dose because of prior neurotoxicity in
children who received a lower dose of FA. Neurotoxicity was
prevented without any reported decrease in event-free survival.
We hypothesize that more and greater toxicity (especially neuro-
toxicity) will result from this dangerous dose reduction trend of
folinic acid rescue. Acceptance of the hypothesis as stated here
will allow HDMTX to be given effectively without causing
unnecessary harm especially brain damage.
This hypothesis can be proved by performing a randomization
in a prospective study of one of the large leukemia study groups
to show that increased dosage of folinic acid can prevent neurotox-
icity without compromising cure.
Conflict of interest statement
None.
References
[1] Larsen EC, Salzer WL, Devidas M, et al. Comparison of high dose methotrexate
(HD-MTX) with Capizzi methotrexate plus aspiraginase (CMTX/ASNase) in
children and young adults with high risk acute lymphoblastic leukemia (HR-
ALL): a report of the Childrens Oncology Study AALL0232. J Clin Oncol
2011;29:2011 (suppl:abst3).
[2] Hegyi M, Gulacsi A, Csagoly E, et al. Clinical relations of methotrexate
pharmokinetics in the treatment of pediatric osteosarcoma. J Cancer Res Clin
Oncol 2012. http://dx.doi.org/10.1007/S 00432-012-1214-2.
[3] Allen JC, Walker R, Rosen G. Preradiation high dose intravenous methotrexate
with leucovorin rescue for untreated primary childhood brain tumors. J Clin
Oncol 1988;6:649–53.
[4] Mahoney DH, Shuster J, Nitshe R, et al. Intermediate-dose intravenous
methotrexate with intravenous mercaptopurine is superior to repetitive low-
dose methotrexate with intravenous mercaptopurine for children with lower-
risk B lineage acute lymphoblastic leukemia: Pediatric Oncology Group III
Phase Trial. J Clin Oncol 1998;16:246–54.
[5] Bonda-Shkedi E, Weyl Ben Arush M, Kaplinsky C, et al. The Correlation
between dose of folinic acid and neurotoxicity in children and adolescents
treated for osteosarcoma with high dose methotrexate (HDMTX): a
Neuropsychological and Psychosocial Study. J Ped Hematol Oncol
2013;35:271–5.
[6] Borsi JD, Wesenberg F, Stokland T, et al. How much is too much? Folic acid
rescue dose in children with acute lymphatic leukemia. Eur J Cancer
1991;27:1006–9.
[7] Cohen IJ. Defining the dose of folinic acid after high dose methotrexate for
acute lymphatic leukemia that will prevent neurotoxicity without rescuing
malignant cells in the central nervous system. J Pediatr Hematol Oncol
2004;26:156–63.
[8] Sterba J, Dusek L, Demlova R, et al. Pretreartment plasma folate modulates the
pharmacodynamic effect of high dose methotrexate in children with acute
lymphoblastic leukemia and non-Hodgkins lymphoma: ‘‘folate overrescue’’
revisited. Clin Chem 2006;52:692–700.
[9] Sirotnak FM, Moccio DM, Dorick DM, et al. Optimization of high-dose
methotrexate with leucovorin rescue therapy in the L1210 leukemia and
sarcoma 180 murine tumor models. Cancer Res 1978;38:345–53.
[10] Kamen BA. Folate and antifolate pharmacology. Semin Oncol 1997;5(Suppl.
18):518–39.
[11] Bertino JR. Rescue techniques in cancer chemotherapy: use of leucovorin and
other rescue agents after methotrexate treatment. Semin Oncol
1977;4:203–15.
[12] Mahoney DH, Shuster JJ, Nitschke R, et al. Acute neurotoxicity in children with
B-precursor acute lymphatic leukemia: an association with intermediate-dose
intravenous methotrexate and intrathecal triple therapy – a pediatric oncology
group study. J Clin Oncol 1998;16:1712–22.
[13] Mantadakis E, Smith AK, Kamen BA, et al. Ratio of methotrexate to folate
uptake by lymphoblasts in children with B-lineage acute lymphoblastic
leukemia: a pilot study. J Pediatric Hematol/Oncol 2000;22:221–6.
[14] Pinedo H, Zaharo DS, Bull JM, et al. The reversal of methotrexate cytotoxicity to
mouse bone marrow cells by leucovorin and nucleosides. Cancer Res
1976;35:4418–24.
[15] Boarman DM, Baram J, Allegra CJ. Mechanism of leucovorin reversal of
methotrexate cytoxicity in human MCF-7 breast cancer cells. Biochem
Pharmacol 1990;40:2651–60.
[16] Koizumi S, Ueno Y, Ohno I, et al. Reversal of methotrexate cytotoxicity to
human bone marrow cells and leukemic K562 cells by
leucovorin:methotrexate polyglutamates formation as a possible
important factor Jap. J Cancer Res 1990;81:1162–7.
[17] Bleyer WA. The clinical pharmacology of methotrexate new applications for an
old drug. Cancer 1978;41:36–51.
[18] Cohen IJ. Progression of osteosarcoma after high dose methotrexate: over
rescue by folinic acid. Pediatr Hematol Oncol 2003;20:579–81.
[19] Cohen IJ. High-dose methotrexate is effective in osteosarcoma. So what is the
problem ? J Pediatr Hematol Oncol 2009;31:892–4.
[20] Sirotnak FM, Donsbach RC, Moccio DM, et al. Biochemical and
pharmacokinetic effects of leucovorin after high dose methotrexate in a
murine leukemia model. Cancer research 1976;36:4679–86.
[21] Seidel H, Moe PJ, Nygaard R, et al. Evaluation of severe adverse events in
patients treated with protocols including methotrexate infusions. Pediatr
Hematol Oncol 1994;11:165–72.
[22] Browman GP, Goodyear M, Archibold S, et al. Modulation of the antitumor
effect of methotrexate by low dose leucovorin in squamous cell head and neck
cancer. J Clin Oncol 1990;8:203–8.
[23] Skarby T, Anderson H, Heldrup J, et al. High leucovorin doses during high-dose
methotrexate treatment may reduce the cure rate in childhood acute
lymphatic leukemia. Leukemia 2006;20:1955–62.
[24] Wolfrom C, Hartmann R, Fengler R, et al. Randomized comparison of 36-hour
intermediate dose versus 4 hour high dose infusion for remission induction in
relapsed childhood acute lymphatic leukemia. J Clin Oncol 1993;11:827–33.
[25] Graf N, Jost W, Muller J, et al. The effect ofmethotrexate pharmokinetics and of
leucovorin rescue on the prognosis of osteosarcoma. Klinische Padiatrie
1990;202:340–6.
[26] Crews KR, Lui T, Rodriguez-Galido C. Et al. high dose methotrexate
pharmokinetics and outcome in children and young adults with
osteosarcoma. Cancer 2004;100:1724–33.
[27] Janka GE, Winkler K, et al. Early intensification therapy in high risk childhood
acute lymphatic Leukemia: lack of benefit from high dose methotrexate.
Haematol Blood Transfus 1987;30:456–60.
[28] Mantadakis E, Cole P, Kamen B. High dose methotrexate in acute lymphatic
leukemia: where is the evidence for its continued use? J Clin Oncol
1992;10:1579–91.
[29] Mahoney DH, Nitschke R, Lauer SJ, et al. Neurotoxicity (NT) in children with
acute lymphatic leukemia (ALL) receiving intensive methotrexate schedules. A
pediatric oncology group study. Proc Am Soc Oncol 1996;15:366 (abstr 1082).
 I.J. Cohen / Medical Hypotheses 81 (2013) 942–947
[30] Kamen B, Winick N, Holcenberg JS. Oral versus intravenous methotrexate:
another opinion. J Clin Oncol 1998;16:2283–4.
[31] Winick N, Shuster JJ, Bowman WP, et al. Intensive oral methotrexate protects
against lymphoid marrow relapse in childhood B-precursor acute lymphatic
leukemia. J Clin Oncol 1996;14:2803–11.
[32] Bertino JD. Leucovorin rescue revisited. J Clin Oncol 1990;8:193–5.
[33] Kamen BA, Vietti T. Evidence of CSF accumulation of 5-methyltetrahydrofolate
during repeated courses of methotrexate plus folic acid rescue. Br J Cancer
1989;60:799.
[34] Mahoney DH. Oral versus intravenous methotrexate: another opinion. J Clin
Oncol 1998;16:2284–5.
[35] Salzer W, Winick NJ, wacker P, et al. Plasma methotrexate, red blood cell
methotrexate, and red blood cell folate values and outcome in children with
precursor B-cell lymphatic leukemia: a report from the children’s oncology
group. J Pediatr Hematol Oncol 2012;34:e1–7.
[36] Joannon P, Oviedo I, Campbell J, Tordecilla J. High-dose methotrexate therapy
of childhood acute lymphoblastic leukemia:lack of relation between serum
methotrexate concentration and creatinine clearance Ped Blood. Cancer
2004;43:17–22.
[37] Jurgens H, Janka G, Ibrahim M, et al. Prognostic significance of exposure to
intermediate-dose methotrexate in children with standard risks ALL: the
COALL 82/85 experience. Haematol Blood Transfus 1992;34:338–42.
[38] Winick NJ, Bowman WP, Kamen BA, et al. Unexpected acute neurological
toxicity in the treatment of children with acute lymphatic leukemia. J Nat
Cancer 1992;84:252–6.
[39] Thyss A, Milano G, Etienne MC, et al. Evidence for CSF accumulation of 5-
methyltetrafolate during repeated courses of methotrexate plus folinic acid
rescue. Br J Cancer 1989;59:627–30.
[40] Niemeyer CM, Gelber RD, Tarbell NJ, et al. Low-dose versus high-dose
methotrexate during remission induction in childhood acute lymphatic
leukemia (protocol 81-01 update). Blood 1991;78:2514–9.
[41] Tishler M, Caspi D, Fishel B, Yaron M. The effects of leucovorin (folinic acid) on
methotrexate therapy in rheumatoid arthritis. Arthritis Rheum
1988;31:906–8.
[42] Hills RJ, Ive FA. Folinic acid rescue used routinely in psoriatic patients with
known methotrexate sensitivity. Acta Derm Venereol (Stockh)
1992;72:438–40.
[43] Cipiano AP, Selsky LM, Bertino JR. Failure of leucovorin ‘‘rescue’’ in
methotrexate treatment of psoriasis. Arch Derm 1970;101:651–2.
[44] Takacs P, Rodriguez L. High folic acid levels and failure of single-dose
methotrexate treatment in ectopic pregnancy. Int J Gynecol Obstet
2005;89:301–2.
[45] Valik D, Sterba J, Bajciova V, et al. Severe encephalopathy induced by the first
but not second course of high dose methotrexate mirrored by plasma
homocysteine elevations and preceded by extreme differences in
pretreatment plaama folate. Oncology 2005;69:269–72.
[46] Allen J, Rosen F, Juergens H, et al. The inability of oral leucovorin to elevate CSF
5-methyl-tetrahydrofolate following high dose intravenous methotrexate
therapy. J Neuro-Oncol 1983;1:39–44.
947
[47] Mehta BP, Glass JP, Shapiro WR. Serum and cerebrospinal fluid distribution of
5-methyltetrahydrofolate after intravenous calcium leucovorin and intra-
Ommaya methotrexate administration in patients with meningeal
carcinomatosis. Cancer Res 1963;43:435–8.
[48] Mantadakis E, Cole PD, Kame B. High dose methotrexate in acute lymphatic
leukemia: where is the evidence for its continued use? Pharmacotherapy
2005;25:748–55.
[49] Reiter A, Schrappe M, Ludwig W-D, et al. Chemotherapy in 998 unselected
childhood acue lymphoblastic leukemia patients. Results and conclusions of
the multicentral trial ALL-BFM 86. Blood 1994;84:3122–33.
[50] Conter V, Arico M, Valsecchi MG, et al. Extended intrathecal methotrexate may
replace cranial irradiation for prevention of CNS relapse in children with
intermediate -risk acute lymphoblastic leukemia treated with Berlin–
Frankfurt–Munster-based intensive chemotherapy. J of Clin Oncol
1995;13:2497–502.
[51] Donadieu J, Auclerc M-F, Baruchel A, et al. Critical study of prognostic factors
in childhood acute lymphatic leukemia: differences in outcome are poorly
explained by the most significant prognostic variables. Brit J Haematol
1998;102:729–39.
[52] Patte C, Auperin A, Michon J, et al. The Societe Francais d’oncologie Pediatrique
LMB89 protocol:highly effective multiagent chemotherapy tailored to thr
tumor burden and initial response in 561 unselected children with B-cell
lymphomas and L3 leukemia. Blood 2001;97:3370–9.
[53] Pui C-H, Relling M, Evans WE. Is megadose of methotrexate beneficial to
patients with acute lymphoblastic lymphoma? Leukemia Lymphoma
2006;47:2431–2.
[54] Goldin A, Venditti JM, Kline, et al. Eradication of leukemic cells (L1210) by
methotrexate plus citrovorum factor. Nature 1966;212:1548–50.
[55] Vezmar S, Becker A, Bode U, Jaehde U. Biochemical and clinical aspects of
methotrexate neurotoxicity. Chemotherapy 2003;49:92–104.
[56] Vezmar S, Becker A, Bode U, Jaehde U. Methotrexate-associated neurotoxicity
is not only insufficient leucovorin rescue. Pediatr Blood Cancer
2009;53:127–8.
[57] Sterba J, Valik D, Bajciova V, et al. High dose methotrexate and/or leucovorin
rescue for the treatment of children with lymphoblastic malignancies: do we
really know why, when, and how? Neoplasma 2005;52:456–63.
[58] Hill F, Vargha-Khadaem F, et al. UKALLXI randomized trial stratified CNS
treatments and prospective neurophysiological assessment in childhood acute
lymphatic leukemia (ALL). Blood 2000;96:466a (Abstract).
[59] Kamen B, Weitman SD. High dose methotrexate is it warrented? Ped Hematol
Oncol 1994;11:135–7.
[60] Kamen B. High dose methotrexate and aspiraginase for the treatment of
children with acute lymphatic leukemia; why and how?J. Pediatr Hematol
Oncol 2004;26:333–5.
[61] Cohen IJ. Methotrexate-associated alterations of the folate and methyl-
transfer pathway in the CSF of patients with and without symptoms of
neurotoxicity. Pediatr Blood Cancer 2009;53:127–8.