Trimetrexate as biochemical modulator of 5-fluorouracil/leucovorin in advanced colorectal cancer: final results of a randomised European study

C. J. A. Punt1,+, H. J. Keizer2,§, J. Douma3, T. Skovsgaard4, J. Schüller5, E. W. Muller6, C. H. H. Ten Napel7, J. J. Croles8, H. Lochs9, J. Zhang10 and L. Hammershaimb10

1University Medical Centre, St. Radboud, Nijmegen; 2University Medical Centre, Leiden; 3Rijnstate Hospital, Arnhem, The Netherlands; 4Herlev Hospital, University Hospital of Copenhagen, Denmark; 5Rudolf Hospital, Vienna, Austria; 6Slingeland Hospital, Doetinchem; 7Medisch Spectrum, Twente Enschede; 8Bosch Medicentrums, Hertogenbosch, The Netherlands; 9University Hospital Charite, Berlin, Germany; 10MedImmune Oncology, Inc., Gaithersburg, MD, USA

Received 7 September 2001; accepted 11 September 2001.


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background

Trimetrexate (TMTX) is a biochemical modulator of 5-fluorouracil (5-FU) and leucovorin (LV). Phase II trials have shown promising activity of 5-FU/LV/TMTX in patients with advanced colorectal cancer (ACC). This trial evaluated the effect of TMTX in combination with 5-FU/LV as first-line treatment in ACC.

Patients and methods

Patients with ACC were randomised to receive either intravenous LV 200 mg/m2/5-FU 600 mg/m2 or TMTX 110 mg/m2 followed 24 h later by LV 200 mg/m2/5-FU 500 mg/m2 plus oral LV rescue. Both schedules were given weekly for 6 weeks every 8 weeks. Patients were evaluated for progression-free survival (PFS), overall survival (OS), tumour response, quality of life (QoL) and toxicity.

Results

A total of 365 patients were randomised. A statistically significant prolongation of median PFS was seen in patients treated with TMTX/5-FU/LV compared with 5-FU/LV (5.4 months versus 4.1 months, respectively; P = 0.03), and a trend towards a significant benefit for OS (13.4 months versus 10.5 months, respectively; P = 0.08). Tumour response, QoL and toxicity were comparable between the two arms. Diarrhoea was the most frequently occurring grade 3 or 4 toxicity (22% and 30%, respectively).

Conclusions

The addition of TMTX to a weekly regimen of 5-FU/LV results in a small but significant improvement in PFS without adding toxicity or worsening QoL in patients with ACC.

Key words: biochemical modulation, colorectal cancer, 5-fluorouracil, leucovorin, trimetrexate, randomised trial


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Trimetrexate (TMTX) is a non-classical folate antagonist. As a single agent it has some activity against a variety of solid tumours [1]. Furthermore, it is a biochemical modulator of 5-fluorouracil (5-FU) and leucovorin (LV) [2]. This biomodulation results from inhibition of the enzyme dihydrofolate reductase, and from the increase of intracellular levels of 5-phosphoribosyl-1-pyrophosphate via inhibition of purine synthesis. This leads to an increased formation of the active 5-FU metabolite, FdUMP. LV, the best-known biomodulator of 5-FU that enhances the complex formation between thymidylate synthase and FdUMP, exerts its modulating activity independently from TMTX. In contrast to methotrexate (MTX), TMTX has a clear synergistic effect in combination with 5-FU and LV [3]. This is explained by the fact that TMTX does not compete with LV for cellular uptake and metabolism: TMTX does not need active transport but enters the cell by passive diffusion, and TMTX does not need to undergo polyglutamation. Based on these findings clinical studies were initiated.

In a study with escalating doses of TMTX and 5-FU in combination with LV, Conti et al. [4] showed that the maximum tolerated dose of a weekly schedule of TMTX/LV/5-FU was TMTX at 110 mg/m2 followed 24 h later by LV and 5-FU, both at 500 mg/m2, and oral LV rescue starting 6 h after 5-FU at 10 mg/m2 every 6 h for seven doses. Treatment was given weekly six times followed by a 2-week rest period. The dose-limiting toxicity was diarrhoea. Also, hypersensitivity reactions to TMTX were observed in several patients. A partial response occurred in seven of 35 pretreated patients (20%) with advanced colorectal cancer (ACC) and in several patients with other gastrointestinal malignancies as well. This phase I schedule was then tested in phase II studies with the only differences being that the dose of intravenous (i.v.) LV was decreased to 200 mg/m2 and the dose of oral LV was fixed at 15 mg. Two phase II studies in previously untreated patients with ACC have been performed. Blanke et al. [5] observed a response rate of 50% in 36 evaluable patients (95% confidence interval (CI) 32% to 68%), with a median overall survival (OS) of 53 weeks. Analysis by intention-to-treat showed a response rate of 42% (95% CI 26% to 58%). The major toxicity was grade 3 or 4 diarrhoea, which occurred in 58% of patients. Although haematological toxicity was infrequent, two patients died from sepsis. Szelényi et al. [6] found a partial response rate of 36% (95% CI 25% to 49%), and an incidence of severe diarrhoea of 22%, decreasing to 8% with early loperamide treatment in 34 patients. Based on these promising results, two independent phase III studies were initiated, one in Europe and one in the United States [7]. In both studies the same schedule of TMTX/5-FU/LV was used as in the phase II studies. Here, we report on the European study, which differs from the USA study in two aspects: (i) it was not placebo-controlled and therefore no oral LV rescue was administered in the control arm, and (ii) the 5-FU dose in the control arm was higher compared with that in the TMTX-containing arm (600 mg/m2 versus 500 mg/m2, respectively). This 5-FU dose has commonly been used in weekly 5-FU/LV regimens in randomised trials [8, 9]. Therefore, in contrast to the USA study, a higher 5-FU dose was used in the control arm, reflecting the intention to investigate a meaningful biomodulation effect of TMTX.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Eligibility criteria included histological evidence of ACC not amenable to curative surgery, bidimensional or unidimensional measurable disease, at least one parameter with a diameter of 2 cm or greater, no prior treatment for advanced disease, prior adjuvant treatment allowed provided 1 year had elapsed since completion, Eastern Cooperative Oncology Group performance status (PS) of 0–2, adequate bone marrow/liver/renal function, and written informed consent.

The primary objective of the study was to detect a difference of a minimum of 3 months in progression-free survival (PFS), with a power of 79% and {alpha} = 0.05. An interim analysis on PFS and toxicity was planned for the European study after the first 200 patients were followed for a minimum of 6 months [10]. Secondary objectives were response rate, response duration, OS, and quality of life (QoL). Stratification was performed for treatment centre, PS (0, 1 or 2), site of primary (colon or rectum), site of metastases (liver dominant with <30% involvement, or liver dominant >30%, or lung dominant or other dominant), disease evaluability (measurable or evaluable), prior adjuvant treatment (yes or no), and serum LDH value (<=300 U/l or >300 U/l).

Patients were randomised to either LV 200 mg/m2 as a 1-h i.v. infusion followed directly by 5-FU 600 mg/m2 as an iv bolus (control arm) or TMTX 110 mg/m2 as a 1-h i.v. infusion followed 22–24 h later by LV 200 mg/m2 as a 1-h i.v. infusion followed directly by 5-FU 500 mg/m2 as an i.v. bolus and LV 15 mg orally every 6 h for seven doses, starting 6 h after 5-FU administration (experimental arm). At the onset of any diarrhoea, patients were instructed to use an intensive dosing schedule of loperamide: 4 mg after the first liquid stool and 2 mg every 2 h until 12 h after the last passage of liquid stool up to a maximum of 48 h. Treatment was given for 6 weeks every 8 weeks. Treatment was continued until disease progression with a maximum of 1 year. Patients were followed weekly for toxicity and every 8 weeks for tumour response. Computed tomography (CT) scans were required for the evaluation of liver metastases. National Cancer Institute Common Toxicity Criteria for toxicity and WHO criteria for tumour response were used. Dose reductions were performed as follows: for haematological toxicity grade 2, 5-FU was reduced to 75%; for haematological and non-haematological toxicity grade 3 or 4, treatment was withheld for a maximum of 2 weeks until recovery to grade 1 or less and then resumed with 5-FU at 50% of the dose. If at 50% dose reduction of 5-FU no toxicities greater than grade 2 occurred, increases in 5-FU by 25% increments were allowed. If treatment was withheld for 3 weeks (not including rest weeks), the patient was removed from the study. Patients were asked to complete the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (QLQ-C30) at baseline and at week 5 and 8 of each cycle and every 3 months during follow-up. Information on any subsequent antitumour treatment was collected in the case report forms. An independent review of CT scans was planned.

This study was conducted in accordance with the principles of the Declaration of Helsinki, as adopted by the 29th World Medical Assembly, Helsinki, Finland, and revised at the 48th World Medical Assembly in Somerset West, Republic of South Africa, 1996. These principles are consistent with those set forth in the International Conference on Harmon-isation Guidelines on Good Clinical Practice, and the current USA Code of Federal Regulations (21 CFR Parts 56 and 50) regarding the requirements for independent ethics committees and institutional review boards and protection of the rights and welfare of human subjects involved in clinical investigations.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
A total of 365 patients from 22 centres throughout Europe were randomised between March 1996 and February 1999; the last patient completed treatment in February 2000. One patient was excluded from the intention-to-treat analysis because of incorrect histological diagnosis. Patient characteristics were well balanced over the two treatment arms, with no statistically significant differences for known prognostic parameters (Table 1). Median follow-up at the time of analysis was 37 months. Patients received a median of 2.0 cycles in the control arm and 3.0 cycles in the experimental arm. These numbers did not change when only cycles in which at least five of the six treatment weeks were completed. The median dose intensity for 5-FU was 87% in the control arm and 92% in the experimental arm.


View this table:
[in this window]
[in a new window]
 
Table 1. Patients characteristics (intention-to-treat dataset)
 
Progression-free and overall survival
There was a statistically significant prolongation of PFS (1.3 months) in patients treated with TMTX/5-FU/LV (median of 4.1 months versus 5.4 months, respectively; P = 0.03 based on Wilcoxon test; hazard ratio of 0.87 with 95% CI 0.71–1.08). This is shown graphically in Figure 1. Although the study was not powered to detect a statistically significant difference in OS, we found a trend towards improved median OS (2.9 months) in patients treated with TMTX/5-FU/LV (median of 10.5 months versus 13.4 months, respectively; P = 0.08 based on Wilcoxon test; hazard ratio 0.86 with 95% CI 0.69–1.07). This is shown graphically in Figure 2. The 1-year and 2-year OS rates for the control and experimental arms were 37%/21% and 54%/27%, respectively. There was a statistically significant difference in median OS for patients with a PS of 0 or 1 versus patients with a PS of 2 (median of 12.3 months versus 4.1 months in the control arm, median of 13.9 months versus 3.6 months in the experimental arm).



View larger version (21K):
[in this window]
[in a new window]
 
Figure 1. Progression-free survival.

 


View larger version (19K):
[in this window]
[in a new window]
 
Figure 2. Overall survival.

 
Response
The overall unconfirmed tumour response rate for 5-FU/LV was 25% (95% CI 19% to 32%) and 29% (95% CI 22% to 35%) for TMTX/5-FU/LV. Since tumour evaluation was performed after every 8-week cycle, a confirmatory evaluation 4 weeks after the first documentation of a response was never performed. If only those responses that were confirmed by another evaluation that was performed at or greater than 4 weeks after the first documentation of a response were included, then the overall response rates were similar at 18% (95% CI 13% to 24%) and 19% (95% CI 13% to 24%), respectively. Data on response are summarised in Table 2. The median duration of response was 5.2 months for 5-FU/LV and 4.0 months for TMTX/5-FU/LV. An independent review of CT scans was performed on patients who achieved an objective response to treatment or had stable disease. Approxi-mately 80% of CT scans were available for independent review. Findings from this review showed that the response rates from the independent CT scan review and the assessment made by the investigator were comparable.


View this table:
[in this window]
[in a new window]
 
Table 2. Tumour response. No statistically significant differences in response rate or duration of response were noted.
 
Subsequent antitumour treatment
Since new effective drugs were introduced during the course of the study, investigators were required to document any subsequent treatment. The percentage of patients receiving further systemic therapy did not significantly differ between the study arms (41% in the 5-FU/LV arm and 47% in the TMTX/5-FU/LV arm). This mainly concerned the use of second-line treatment with irinotecan, which was administered less frequently in the control arm (17%) compared with the experimental arm (26%; P = 0.04). Subsequent radiotherapy was given to 13% of patients in both study arms, and surgery in 6% and 7% of patients, respectively.

Toxicity
Grade 3 or 4 toxicities are summarised in Table 3. The most prominent toxicity was diarrhoea, which occurred less frequently in patients treated with TMTX/5-FU/LV. There was an equal use of loperamide in both study arms (46% in the control arm, 45% in the experimental arm). Of note, there was a low incidence of hypersensitivity reactions. Five deaths were probably or certainly treatment-related, three in the TMTX arm and two in the control arm. In the TMTX arm, two deaths were due to neutropenic sepsis and one death was due to heart attack. In the control arm, one death was due to neutropenic sepsis and one death was due to cerebrovascular accident. The percentage of patients who discontinued treatment due to adverse events was 16% in the TMTX arm and 20% in the control arm.


View this table:
[in this window]
[in a new window]
 
Table 3. Treatment-related grade 3 or 4 toxicities (%). No significant differences were found. Five treatment-related deaths occurred in the study, three in the trimetrexate (TMTX) arm and two in the control arm. In the TMTX arm, two deaths were due to neutropenic sepsis and one death was due to heart attack. In the control arm, one death was due to neutropenic sepsis and one death was due to cerebrovascular accident.
 
Quality of life
Compliance on the EORTC QLQ-C30 ranged from 81% to 90% in the experimental arm and 81% to 96% in the control arm through the first five cycles of treatment. Thereafter, the number of patients on-study dropped off markedly. The results of the QoL showed that there were no statistically significant differences in any of the mean EORTC QLQ-C30 functional or symptom scores between the two treatment arms.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Meta-analyses of randomised trials have demonstrated effective biochemical modulation of 5-FU for both LV and MTX [9, 11], and 5-FU/LV still remains the cornerstone of treatment of ACC. Addition of new drugs like irinotecan or oxaliplatin to 5-FU/LV does increase response rates and sometimes survival [12–15], although a rigorous prospective comparison of the concomitant versus the sequential use of these new agents has never been performed.

TMTX also has biochemically modulating effects on 5-FU and has several advantages over MTX. Based on promising phase II results of TMTX/5-FU/LV in ACC [5, 6], two independent randomised studies were initiated, and both were designed to detect a statistically significant difference in PFS of TMTX/5-FU/LV over 5-FU/LV. In our European study, we found an increase in median PFS of 1.3 months. Although this difference was statistically significant with a P value of 0.03 (based on the Wilcoxon test), the clinical relevance of this difference is questionable. We found an increase in median OS of 2.9 months, which showed a trend towards significance (P = 0.08; Wilcoxon test). It should be noted that our study was not designed to detect a statistically significant difference of this magnitude for this end point. An integrated analysis on the OS and PFS of the European and the USA studies has been performed [16]. A possible bias of OS in our study by the more frequent use of second-line irinotecan in the experimental arm cannot be excluded since irinotecan has been shown to increase survival in this setting [17]. The somewhat lower median OS compared with more recent randomised studies using irinotecan [12, 13] or oxaliplatin [14, 15] can be explained by the fact that these agents were not yet commonly available for salvage treatment during the larger part of our study accrual period, and a more unfavourable patient population in our study with almost 50% of our patients with three or more metastatic sites, compared with approximately 10% to 15% in these other studies [12–15].

The high incidence, up to 58%, of severe diarrhoea in previous phase II studies [5, 6] prompted us to prescribe intensive dosing of loperamide at the onset of diarrhoea, which occurred in both treatment arms. This resulted in a marked decrease in the incidence of diarrhoea, as was shown by others [6]. The slightly lower incidence of diarrhoea in the experimental arm was unexpected, and cannot be explained by a difference in the use of loperamide. In general, treatment with TMTX/5-FU/LV was well-tolerated, and the results on toxicity in the 5-FU/LV arm are very comparable with those of another randomised study using this schedule [8].

In conclusion, this study shows that the addition of TMTX to a weekly schedule of 5-FU/LV results in a small but signifi-cant improvement of PFS, with a trend towards a significant benefit in OS. Treatment with TMTX/5-FU/LV is feasible with an acceptable incidence of severe diarrhoea. Given this safety profile, a combination of this schedule with other cytotoxic drugs appears feasible.


    Acknowledgements
 
The following investigators have contributed to this study. In The Netherlands: H. R. Oosten, Canisius Wilhelmina Hospital, Nijmegen; G. Vreugdenhil, St Joseph Hospital, Veldhoven. In Germany: M. Clemens, Akademisches Lehrkrankenhaus, Trier; M. Lorenz, Goethe University Hospital, Frankfurt; H. H. Gerhartz and P. Ritter, Städtische Kliniken, Duisburg; E. D. Kreuser, University Hospital Benjamin Franklin, Berlin; G. Schlimok, Zentral Klinikum, Augsburg; S. Oehl, St Antonius Hospital, Wuppertal; J. Potenberg, EV Waldkrankenhaus Spandau, Berlin. In Austria: J. Schüller, Krankenstalt, Rudolfstiftung, Vienna. In the UK: P. Price, Hammersmith Hospital, London. In France: J. M. Vannetzel, Hartmann Clinic, Paris. In Italy: M. Nardi, Instituto Regina, Elena, Rome. In Belgium: D. Schrijvers University Hospital Antwerp. The authors would like to thank Dr Stephen G. Gwyther, Department of Radiology, East Surrey Hospital, Redhill, Surrey, UK, for his assistance in the independent review of CT scans, and Dr James Balsley, MedImmune Oncology, Inc. Gaithersburg, MD, USA for his assistance in the medical review and analyses of the data.


    Footnotes
 
+ Correspondence to: Department of Medical Oncology, University Medical Centre, St Radboud, PO Box 9101, 6500 HB Nijmegen, The Netherlands. Tel: +31-24-3615215; Fax: +31-24-3540788; E-mail: c.punt@onco.azn.nl Back

§ Deceased. Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Haller DG. Trimetrexate: experience with solid tumors. Semin Oncol 1997; 24: 71–76.

2. Bertino JR. Biomodulation of 5-fluorouracil with antifolates. Semin Oncol 1997; 24 (Suppl 18): 52–56.

3. Romanini A, Li WW, Colofiore JR, Bertino JR. Leucovorin enhances cytotoxicity of trimetrexate/fluorouracil, but not methotrexate/fluorouracil, in CCRF-CEM cells. J Natl Cancer Inst 1992; 84: 1033–1038.[Abstract]

4. Conti JA, Kemeny N, Seiter K et al. Trial of sequential trimetrexate, fluorouracil, and high-dose leucovorin in previously treated patients with gastrointestinal carcinoma. J Clin Oncol 1994; 12: 695–700.[Abstract]

5. Blanke CD, Kasimis B, Schein P et al. Phase II study of trimetrexate, fluorouracil, and leucovorin for advanced colorectal cancer. J Clin Oncol 1997; 15: 915–920.[Abstract]

6. Szelényi H, Hohenberger P, Lochs H et al. Sequential trimetrexate, 5-fluorouracil and folinic acid are effective and well tolerated in metastatic colorectal carcinoma. The phase II study group of AIO. Oncology 2000; 58: 273–279.[ISI][Medline]

7. Blanke CD, Shultz J, Cox J et al. A double-blind placebo-controlled randomized phase III trial of 5-fluorouracil and leucovorin, plus or minus trimetrexate, in previously untreated patients with advanced colorectal cancer. Ann Oncol 2002; 13: 87–91.[Abstract/Free Full Text]

8. Buroker TR, O’Connell MJ, Wieand HS et al. Randomized comparison of 2 schedules of fluorouracil and leucovorin in the treatment of advanced colorectal cancer. J Clin Oncol 1994; 12: 14–20.[Abstract]

9. The Advanced Colorectal Cancer Meta-Analysis Project. Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: evidence in terms of response rate. J Clin Oncol 1992; 10: 896–903.[Abstract]

10. Punt CJA. Trimetrexate as a biochemical modulator of 5-fluorouracil and leucovorin in colorectal cancer. Semin Oncol 2000; 27 (Suppl 10): 88–90.[ISI][Medline]

11. The Advanced Colorectal Cancer Meta-Analysis Project. Meta-analysis of randomized trials testing the biochemical modulation of fluorouracil by methotrexate in metastatic colorectal cancer. J Clin Oncol 1994; 12: 960–969.[Abstract]

12. Saltz L, Cox JV, Blanke C et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. Irinotecan Study Group. N Engl J Med 2000; 343: 905–914.[Abstract/Free Full Text]

13. Douillard JY, Cunningham D, Roth AD et al. Irinotecan combined with fluorouracil compared with fluorouracil alone as first-line treatment for metastatic colorectal cancer: a multicenter randomised trial. Lancet 2000; 355: 1041–1047.[ISI][Medline]

14. Giacchetti S, Perpoint B, Zidani R et al. Phase III multicenter randomized trial of oxaliplatin added to chronomodulated fluorouracil-leucovorin as first-line treatment of metastatic colorectal cancer. J Clin Oncol 2000; 18: 136–147.[Abstract/Free Full Text]

15. de Gramont A, Figer A, Seymour M et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000; 18: 2938–2947.[Abstract/Free Full Text]

16. Punt CJ, Blanke C, Zhang J, Hammershaimb L. Integrated analysis on overall survival of two randomised studies comparing 5-fluorouracil/leucovorin with or without trimetrexate in advanced colorectal cancer. Ann Oncol 2002; 13: 92–94.[Abstract/Free Full Text]

17. Cunningham D, Pyrhönen S, James RD et al. Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure for patients with metastatic colorectal cancer. Lancet 1998; 352: 1413–1418.[ISI][Medline]