Figure 1. Treatment Scheme
Cancer Therapy Vol 1 203-208, 2003.
Phase I study of high dose
5-fluorouracil and folinic acid in weekly continuous infusions
Limacher JM*,
Duclos B, Dufour P, Wihlm J€, Leveque D , Jehl FŠ,
Eichler F, Keiling R, Natarajan-Ame S, Kurtz JE, Bergerat JP
Dpartement
dŐHmatologie et dŐOncologie, Service de Pharmacie, Hpitaux
Universitaires de Strasbourg, 1 place de lŐHpital, 67091 Strasbourg Cedex,
France, €Laboratoire
de Biochimie, Centre Rgional de Lutte Contre le Cancer Paul Strauss, Strasbourg, ŠLaboratoire de Bactriologie, Facult de Mdecine, Strasbourg
__________________________________________________________________________________
*Correspondence: Jean-Marc Limacher, Dpartement d'Hmatologie et
d'Oncologie, Hpitaux Universitaires de Strasbourg, 1 place de l'Hpital, 67091
Strasbourg Cedex; Tel: 03.88.11.57.85; Fax: 03.88.11.63.60; e-mail:
Jean-Marc.Limacher@chru-strasbourg.fr
Key Words: 5-fluorouracil (5FU),
folinic acid, chemotherapy, colorectal cancer, breast cancer, hand-foot
syndrome
Summary
We present a
phase I study of a 5-fluorouracil (5FU) and folinic acid combination given at
high doses in weekly continuous infusions. The aim of the study was to
determine the maximal tolerated dose of 5FU and the encountered toxicities. 5FU
was given as a weekly 24 hr continuous infusion at doses between 1.8 and 2.7
g/m2. 600 mg/m2 of calcium folinate was given as a 200 mg
loading dose followed by the rest as a continuous infusion simultaneously with
5FU. All 42 patients entered in the study were diagnosed with malignant
diseases in which 5FU chemotherapy was of potential benefit. Plasma
concentrations of 5FU and folinic acid during the continuous infusions were
determined in three of them. The treatment was well tolerated using up to 2.4
g/m2 of 5FU with only minor side effects. At higher doses (2.5, 2.6
and 2.7 g/m2) the toxic manifestations became rapidly more
important. Diarrhea, nausea and vomiting and hand-foot syndrome were the most
frequent. Other toxicities observed included: angina pectoris, transient
encephalopathy and colectasia. The loading dose of calcium folinate allowed
plasma concentrations of reduced folinate compatible with optimal potentiation
of 5FU during nearly the entire time of the infusion. The maximal tolerated
dose of 5FU under such conditions is 2.4 g/m2. Up to this
dosage the treatment can be used in even heavily pretreated patients with
conservation of quality of life. The therapeutic value of this treatment is
currently evaluated in its principal indication, advanced colorectal cancer, in
comparison with the classical five day bolus administration.
Though
new drugs like oxaliplatin or irinotecan have improved the treatment of
colorectal cancer they are usually associated to 5-fluorouracil (5FU) which
remains the pivotal drug in this pathology (Baker, 2003; Rougier and Mitry, 2001). Nevertheless, the best way of
administering 5FU in this indication remains a matter of debate (Leichman et al, 1995). Experimental and clinical
studies have found continuous infusions to give higher response rates compared
to bolus injections (Seifert et al, 1975; Lokich et al, 1989; Weinermann et al,
1990; Patel et al, 2003). A dose effect for 5FU has clearly been demonstrated
for both bolus and continuous infusions, yet the latter allows the delivery of
higher dose intensities (Hrynuk et al, 1987). The other way of improving 5FU
activity is by biomodulation. Several drugs have been associated with 5FU but
only folinic acid is considered to increase response rates and survival (Khne
Wmpner et al, 1992; Piedbois and Michiels 2003). The 5FU-folinic acid
association is usually given in a five day schedule using bolus administrations
repeated every three to four weeks (Machover et al, 1986). A different way of
administering this combination has been described based on a simultaneous
continuous 24hr infusions of the two compounds. 5FU was given at 2.6 g/m2
with folinic acid at 500 mg/m2. Despite the high dose intensity of
5FU the toxic side effects were limited. Its efficacy in the treatment of
advanced colorectal cancers was interesting with an overall response rate of 45
% reaching 52% in
previously untreated patients (Ardalan et al, 1991). We present a phase I study
of a similar treatment modified by the addition of a loading dose of folinic
acid before the continuous infusions (Figure 1). The goal of the study was to determine the
maximal tolerated doses of 5FU under such conditions and to describe the
encountered toxicities.
Figure 1. Treatment Scheme
5FU
(Roche) was given as a 24 hr continuous infusion, the drug being diluted in 5%
glucose
and administered by a portable pump (CADD, Pharmacia Deltec) connected to a
sub-cutaneously implanted double lumen venous access (Porth-a-cath, Pharmacia).
The dose of 5FU for the first patients was 1.8 g/m2. The explored
dose stages were 1.8, 2.0, 2.2, 2.4, 2.5, 2.6 and 2.7 g/m2. At least
three patients were treated at each dose level. If no major toxicity was
observed at a particular dose it was then increased for the next group of
patients. For a given patient, the dose of 5FU administered weekly remained
constant throughout the study.
The
calcium folinate dose was 600 mg/m2. Two hundred mg was administered
as a two hours infusion with the rest of the dose delivered over 24 hours
concommitanly with 5FU using another portable pump (Singleday Infusor 2 ml/hr,
Baxter) connected to the implanted port. Calcium folinate and 5FU are not
stable when mixed together for several hours and therefore cannot be placed
together. This treatment was repeated every week until progression. In the
event of grade 2 or 3 toxicity the treatment was delayed until resolution and
the 5FU dose subsequently decreased by 0.2 g/m2.
All
patients underwent a weekly clinical examination assessing tolerance to the
treatment. Toxic events were graded according to WHO criteria. Whenever
possible the response to treatment was assessed every two months.
5FU and
folinic acid plasma levels during continuous infusion were measured in three
patients at the 1.8 g/m2 dose level. Five ml blood samples were
taken every three hours during the infusion. Blood was immediately centrifuged
at 3000 rpm for 5 min
and 2 ml of the plasma were removed and frozen at -200C.
This treatment was administered to informed and consenting patients suffering from a malignant disease in which chemotherapy with 5FU was indicated. Other eligibility criteria were age less than 70 years, ECOG performance status 0, 1 or 2, WBC > 4000/mm3, platelets >100,000/mm3, and no major hepatic or renal dysfunction.
5FU was measured in
the serum by HPLC as previously described (Christophidis et al, 1979). The
limit of detection was 5 ng/ml. The within-run and day to day precision
expressed in terms of coefficient of variation was less then 5%.
D. Determination of plasma folinic acid concentration
Folinic acid was measured in the plasma
by a specific HPLC method. Briefly, 500 ”l of plasma were deproteinized with
500 ”l of acetonitrile. After centrifugation, 3.5 ml of methylene chloride were
added to the supernatant. The mixture was gently shaken by rotation for 10 mn
(20 rpm) and then centrifuged for 10 mn at 1000 g. Twenty microlitres of the
upper aqueous layer were injected in the chromatograph. The chromatograph
consisted of a 126 programmable solvent delivery module (Beckman, Fullerton,
CA), a model 210 sample injection valve with a 20 ”l 100P (Beckman), and a
model 166 programmable wavelength detector (Beckman). Chromatograms were
processed by a GOLD chromatographic data system (Beckman). The assay was
carried out using a 125 x 4.6 mm (inner diameter) C18 column (5 ”m particle size, Merck, Darmstadt, Germany). The mobile phase
consisted of 8% acetonitrile in 5mM tetrabutylammonium bromide (final
concentration) adjusted to pH 2.0 and delivered at a flow rate of 1 ml/mn.
Quantization was based on UV detection at 280 nm. The limit of detection in
plasma was 200 ng/ml. Within- and between-day reproducibilities were less than
10% in terms of coefficient of variation.
A. Patients
42
patients entered the study during the 2 year study period. Population
characteristics are summarized in Table 1.
B.
Toxicity
The
administration of 5FU doses of between 1.8 and 2.4 g/m2 was very
well tolerated with only occasionally grade I minor side effects including mild
erythrodermatitis, nausea and diarrhea. For the three higher dose levels of 5FU
(2.5, 2.6 and 2.7 g/m2)
the toxicities became quickly more frequent and more intense. The most
frequently observed toxic effects were hand-foot syndrome, nausea/vomiting and
diarrhea (Figure 2).
The
hand-foot syndrome was present after three to five weeks treatment in about
half of the patients receiving 2.5 g/m2 or more 5FU. Hand-foot
syndrome was treated with moisturizing creams and dose reduction. Grade II or
III diarrhea and nausea/vomiting were observed in 50% and 75% of the patients who
received 2.6 and 2.7 g/m2 respectively.
Table
1. Patients characteristics
_________________________________________________________________________
Number of
patients: 42
Mean age: 56
(38-76)
Men: 16
Women: 26
Previously treated: 28
Untreated: 14
Distribution
according to pathology:
Colorectal cancer: 23
Breast cancer: 9
Ovarian cancer: 3
Gastric cancer: 2
Hepato-biliary cancer: 2
Soft tissue sarcoma: 2
Bladder cancer: 1
Distribution
according to 5FU dose:
1.8 gr/sqm 11
2.0 gr/sqm 3
2.2 gr/sqm 3
2.4 gr/sqm 5
2.5 gr/sqm 8
2.6 gr/sqm 8
2.7 gr/sqm 4
_________________________________________________________________________
Figure 2: Frequency per dose
level of the three most frequent toxicities (% patients with WHO grade II or
higher toxicities)
Diarrhea
disappeared usually after two weeks of rest and dose reduction. Nausea and
vomiting when present were observed during the two first days of the cycle.
Minor nausea was treated with metoclopramide, more severe nausea and vomiting
responded to 8 mg ondansetron or 3 mg granisetron given intravenously before
the start of the 5FU infusion. Beside these main side effects we observed in
two patients an episode of colectasia with ileus. The two patients had in
common their age of more then 65 years and a dose of 5FU of 2.4 g/m2
or more. In one case clostridium difficile toxin was detected in the
stool. The treatment applied was diet and vegetal coal, whreas colonoscopic
exsuflation was necessary in one patient. Two patients presented a grade II
mucositis at 5FU doses of 2.5 and 2.6 g/m2. Angina pectoris was
observed during 5FU infusion in one patient. This patient had a previous
history of coronary disease. This cardiac complication appeared during the
first cycle and the treatment was not repeated. A thirty year old woman
presented a neurological disorder with sleepiness and confusion spontaneously
regressive within 24 hr, during the first infusion (5FU 2.5 g/m2).
The cerebral CT Scan was normal.
Some patients reported watery eyes; slight alopecia
was present in a few patients after several months of treatment. Grade III
neutropenia has been observed in one 70 years old patient treated with 2.6 gr/m2
of 5FU. No other haematological or biological grade II or higher toxicity was
noted. Neither toxic death nor grade IV toxicity were encountered.
C. Responses to treatment
The response to treatment was not the goal of the
study but was evaluated whenever possible. The responses observed in the
patients with colorectal or breast cancers are presented in Table 2.
The objective response rate in previously untreated
patients with advanced colorectal cancer (3/7) is consistent with the response
rate observed by Ardalan et al, (1991). No patient previously treated for
metastatic colorectal cancer achieved an objective response. One of the
patients with breast cancer achieved a complete response after the following
history: liver metastasis treated by FAC, complete response, intensification
chemotherapy with autologous bone marrow graft, relapse after 7 months, no
response to 5FU-CDDP. She was then treated with 1.8 g/m2 of 5FU. One
women with ovarian cancer refractory to carboplatine based chemotherapy
achieved complete response with the 2.4 g/m2 weekly administration
of 5FU.
D. Pharmacokinetic study
1.
5-fluorouracil plasma levels
The 24 hr continuous infusions of 5FU were started at
3:00 PM. The 5FU plasma concentration increased rapidly in the first six hours,
then more slowly between the sixth and twelfth hour. The maximal
concentrations, 600 to 800 ng/ml (4.5 to 6”M), were observed between 0:00 AM and 3:00 AM.
The concentrations decreased subsequently despite the regular infusion of 5FU (Figure
3). These variations are consistent with the
circadian variations observed by Petit and al (1988).
Table 2: Responses to therapy observed in patients
with colorectal or breast cancer
_________________________________________________________________________
Colorectal cancer
Patients with response
evaluation: 17
Untreated
patients 7
Complete
response: 1
Partial
response: 2
Stabilisation: 4
Progression: 0
Previously
treated patients: 10
Complete response: 0
Partial response 0
Stabilisation: 8
Progression : 2
Breast cancer
Patients with
response evaluation :
6
All pre-treated
Complete response: 2
Partial response: 0
Stabilisation: 2
Progression: 2
_________________________________________________________________________
Figure 3: 5FU Pharmacokinetics
2. Folinic acid plasma levels
After the infusion of 200 mg of calcium folinate over
two hours, the observed plasma concentration of
6[R,S]5,10-methylene-tetra-hydro-folate (THF) were approximately equal to or
above 20 ”M. This concentration has been reported to be mandatory in-vitro to
obtain efficient potentiation of 5FU cytotoxicity (Moran and Scanlon 1991;
Zhang et al, 1992). Subsequently during the continuous infusion of calcium
folinate, the plasma concentrations continued to increase up to between 30 to
50 ”M (Figure 4)
In this
study, we define the maximal tolerated dose of 5FU to be 2.4 g/m2.
At higher doses of 5FU the toxic manifestations are more frequent as described
by Ardalan et al (1991) using 2.6 g/m2. We believe this difference
is due in part to the slightly higher dose of calcium folinate (600 versus 500
mg/m2) and in particular to the use of a 200 mg loading dose. This
loading dose leads to plasma concentrations of 6[R,S]5,10-methylene-THF
compatible with the optimal potentiation of 5FU during nearly all the time of
the continuous 5FU infusion.
The
pattern of the encountered toxicities is in agreement with the known side
effects of 5FU in similar treatment schemes (Nobile et al, 1993). Nausea and
vomiting, even if moderate, are common with 5FU in all the treatment schemes.
Diarrhea is present in continuous infusions and in 5 day treatments. The
hand-foot syndrome is usually noted with the continuous infusions.
Myelosuppression which is common when using repeated daily bolus administration
of 5FU was absent in our series. The possibility of coronary manifestations
during 5FU continuous infusions justifies in our opinion a pre-treatment
cardiac evaluation in patients with a cardiac history, and if necessary the
prophylactic use of calcium channel inhibitors. Encephalopathy is a rare but
known side effect of 5FU. Colectasia has not previously been reported in
patients treated with 5FU and/or calcium folinate. The occurrence of colectasia
in our patients after several courses of treatment suggests a cumulative effect
of the treatment on the bowel but remains without a pathophysiological
explanation.
The
interindividual differences in the tolerance of 5FU are likely to arise from
differences in the metabolism of the drug. The differences observed in the
areas under the curve (AUC) for three patients treated with the same dose of
5FU are consistent with this hypothesis. It raises the possibility of adapting
5FU doses to individual pharmacokinetic parameters. Circadian variations in 5FU
plasma concentration have been previously observed by Petit et al. (1988) and
are the basis of chronotherapy treatment schemes (Lvi et al, 1994). The high
doses of calcium folinate used in this study are justified if we admit that a 6[R,S]5,10-methyiene-THF
plasma concentration higher than 20 ”M is necessary for the optimal
potentiation of 5FU at the level of the thymidylate synthetase. In this
respect, a prospective comparative study has found significantly higher
response rates in advanced colorectal cancer patients treated with 5FU and
high-dose calcium folinate compared with 5FU and low-dose folinate (Jger et
al, 1991). Despite the high dose-intensity of this treatment, the toxic
manifestations are not frequent if the 5FU dose does not exceed 2.4 g/m2.
Administered in this manner, such a chemotherapy schedule can be very useful in
the treatment of advanced tumours like colorectal cancer or other
adenocarcinoma especially if a dose related effect is desired. The lack of
myelosuppression allows its use in patients presenting with poor haematological
reserve or heavy prior chemotherapy. Ambulatory administration which is made
possible by the use of portable pumps preserves quality of life. We are
presently comparing, in a multi-centric randomized study, this treatment with a
classical five day bolus schedule for patients with advanced colorectal cancer.
In conclusion, this phase I study confirmed the feasibility of this treatment
scheme and determined the MTD.
We thank Dr Michael Nord for reviewing of an earlier
draft of the manuscript.
Figure
4: 6[R,S]5,10-methylene-tetra-hydro-folate
pharmacokinetics
Ardalan
B, Chua L, Tian EM, Reddy R, Sridhar K, Benedetto P, Richman S, Legaspi A,
Waldman S, Morrell L, Feun L, Savaraj N and Livingstone A (1991) A phase II study of
weekly 24-hour infusion with high-dose fluorouracil with leucovorin in colorectal
carcinoma. J Clin Oncol 4, 625-630.
Baker DE (2003) Oxaliplatin: a new
drug for the treatment of metastatic carcinoma of the colon or rectum. Rev
Gastroenterol Disord. 3, 31-8.
Christophidis
N, Mihaly G, Vajda E, and Louis W (1979) Comparison of liquid and
gas-liquid chromatographic assays of 5-fluorouracil in plasma. Clin Chem 1, 83-86.
Hryniuk
WM, Figueredo A, and Goodyear M (1987) Applications of dose intensity
to problems in chemotherapy of breast and colon cancer. Semin Oncol 14, 3-11.
Jger E,
Klein O, Bernhard H, Wchter B, Heike M, Theiss F, Dippold W, Meyer zum
Bschenfelde H and Knuth A (1994) Weekly high dose folinic
acid(FA)/5-fluorouracil(FU) versus low dose FA/FU in advanced colorectal
cancer. Results of a randomized multicenter trial. Proc Am Soc Clin Oncol 13, 192 (abstr
556).
Khne
Wmpner CH, Schmoll HJ, Harstrick A, and Rustum YM (1992) Chemotherapeutic
strategies in metastatic colorectal cancer: An overview of current clinical
trials. Semin Oncol 19(2 suppl 3), 105-125.
Leichman
CG, Fleming TR, Muggia FM, Tangen CM, Ardalan B, Doroshow JH, Meyers FJ,
Holcombe RF, Weiss GR, Mangalik A, and Macdonald JS (1995) Phase II study of
fluorouracil and its modulation in advanced colorectal cancer: a Southwest
Oncology Group study. J Clin Onco 13,1303-1311.
Lvi F,
Zinadi R, Di Palma M, Faggiulo R, Garuffi C, Chollet P, Focan C, Iacobelli S,
Perpoint B, Le Rol A, Itzaki M, Vannetzel JM and Misset JL (1994) Improved
therapeutic index through ambulatory circadian rhythmic delivery of high dose
3-drug chemotherapy in a randomized phase III multicenter trial. Proc Am Soc
Clin Oncol 13, 197 (abstr 574).
Lokich
J, Ahlgren JD, Gullo JJ, Philips JA, and Fryer JG (1989) A prospective
randomized comparison of continuous infusion fluorouracil with a conventional
bolus schedule in metastatic colorectal carcinoma: A mid-Atlantic Oncology
Program Study. J Clin Oncol 7, 425-432.
Machover
D, Goldschmidt E, Chollet P, Metzger G, Zittoun J, Marquet J, Vandenbulcke JM,
Misset JL, Schwarzenberg L, Fourtillan JB, Gaget H and Math G (1986) Treatment of
advanced colorectal and gastric adenocarcinomas with 5-fluorouracil and
high-dose folinic acid. J Clin Oncol 4, 685-696.
Moran
RG, Scanlon KL (1991) Schedule-dependent enhancement of the cytotoxicity of
fluoropyrimidines to human carcinoma cells in the presence of folinic acid.
Cancer Res 51, 4618-4623.
Nobile
MT, Sanguineti O, Barzacchi MC, Percivale PL, Bertoglio P, Meszaros PM,
Ardizzoni A and Rosso R (1993) Twenty-four hours weekly infusion of
5-fluorouracil and oral 6S-leucovorin in advanced colorectal cancer: a phase II
study, preliminary results. Proc Am Soc Clin Oncol 12, 227 (abstr
699).
Patel M,
Ardalan K, Hochman I, Tian EM, and Ardalan B (2003) Cytotoxic effects
and mechanisms of an alteration in the dose and duration of 5-fluorouracil.
Anticancer Res 23, 447-452.
Petit E,
Milano G, Lvi F, Thyss A, Bailleul F, Schneider M (1988) Circadian Rhythm-varying
plasma concentration of 5-fluorouracil during a five-day continuous venous
infusion at a constant rate in cancer patients. Cancer Res 48, 1676-1679.
Piedbois
P and Michiels S for the Meta-Analysis Group In Cancer (2003) Survival benefit of
5FU/LV over 5FU bolus in patients with advanced colorectal cancer: An updated
meta-analysis based on 2,751 patients. Proc Am Soc Clin Oncol 22, 294 (abstr
1180).
Rougier P, Mitry E (2001) Review of the role of CPT-11 in the treatment of colorectal cancer. Clin
Colorectal Cancer 1,87-94
Seifert
P, Baker LH, Reed ML, and Vaitkevicius VK (1975) Comparison of continuous
infused 5-fluorouracil with bolus injection in the treatment of patients with
colorectal adenocarcinoma. Cancer 36,123-128.
Weinermann
B, Shah A, Fields A, Kerr I, Cripps C, Shepherd F, Wierzbicki R, Temple W,
Maroun J, Bogues W and Pater J (1990) A randomized trial of
continuous systemic infusion vs bolus therapy with 5-fluorouracil in metastatic
measurable colorectal cancer. Proc Am Soc Clin Oncol 9, 103 (abstr 399).
Zhang
ZG, Harstrick A, Rustum YM (1992) Modulation
of fluoropyrimidines: role of dose and schedule of leucovorin administration. Semin
Oncol 19(suppl 3), 10-15.
Dr. Jean-Marc Limacher