Adjuvant chemotherapy in the treatment of colon cancer: randomized multicenter trial of the Italian National Intergroup of Adjuvant Chemotherapy in Colon Cancer (INTACC)

Original article

Annals of Oncology 14: 1365–1372, 2003 DOI: 10.1093/annonc/mdg359

Adjuvant chemotherapy in the treatment of colon cancer: randomized multicenter trial of the Italian National Intergroup of Adjuvant Chemotherapy in Colon Cancer (INTACC)

1

Gruppo Oncologico Italiano di Ricerca Clinica (GOIRC), Parma; 2Gruppo Cooperativo Istituto Nazionale per la Ricerca sul Cancro (IST-Genova); Gruppo Oncologico Piemontese Tumori Apparato Digerente (GOPTAD), Torino; 4Istituto Oncologico Romagnolo (IOR), Forlì; 5 Gruppo Oncologico del Nord-Ovest (GONO), Pisa; 6Servizio di Epidemiologia Clinica (IST)-Centro Trials (CBA), Genova, Italy 3

Received 12 November 2002; revised 18 March 2003; accepted 11 April 2003

Background: To determine whether the addition of leucovorin to the combination 5-fluorouracil plus levamisole prolongs disease-free survival and overall survival in patients with radically resected colon cancer (Dukes’ B2–3 and C). Patients and methods: Patients (1703) were accrued between March 1992 and February 1995 in a largescale clinical trial within five Italian cooperative groups. After stratification for center, patients were randomized as follows: arm A, 5-fluorouracil [450 mg/m2 intravenous (i.v.) bolus on days 1–5] and levamisole (150 mg orally for 3 days, every 14 days for 6 months) versus arm B, 6-S-leucovorin (100 mg/m2 i.v. bolus on days 1–5) followed by 5-fluorouracil (370 mg/m2 i.v. bolus on days 1–5), plus levamisole (as arm A), every 4 weeks for six cycles. Results: After a median follow-up of 6.4 years no significant difference was seen for either disease-free survival (58% versus 60%, not significant) or 5-year overall survival (68% versus 71%, not significant), respectively. Gastrointestinal toxicity (World Health Organization grade 3/4) was more frequent in arm B (8% versus 18%, not significant). Conclusions: In this trial the schedules used showed no statistically significant differences in terms of diseasefree survival or overall survival in the treatment of colorectal cancer. Key words: adjuvant chemotherapy, colon cancer

Introduction Colon cancer is a major public health problem in Italy and in other Western European countries. In Europe, mortality has declined whilst incidence has increased over the last 20 years [1]. After colon cancer resection, patients with stage II disease have a 5-year survival of ∼75–80%, while in patients with stage III disease 5-year survival rarely exceeds 60%. Approximately 45% of all resected patients have positive lymph nodes and 33% of these will develop recurrent disease [2]. Over the past 30 years, numerous trials have been performed to improve the outcome of patients with operable colon cancer. On the basis of the positive results of the Intergroup Trial 0035 [3], in 1990 a Consensus Development Conference sponsored by the National Institute of Health (NIH) recom-

*Correspondence to: Dr F. Di Costanzo, Unità Operativa Oncologia Medica, Azienda Ospedale Careggi, Via Pieraccini 17, 50139 Firenze, Italy. Tel: +39-055-4277297; Fax: +39-055-4277298; E-mail: [email protected] © 2003 European Society for Medical Oncology

mended 5-fluorouracil (5-FU) plus levamisole (LEVA) as standard adjuvant treatment for stage III colon cancer [4]. In general, most of the studies conducted in the 1990s used the combination of 5-FU and LEVA as the standard against new experimental therapies. One of the potential approaches to increase the efficacy of 5-FU and LEVA was the modulation of 5-FU by leucovorin (LV). LV had been shown to modulate the cytotoxic effect of 5-FU in various in vitro models through the enhancement of the inhibition of thymidylate synthetase (TS) [5]. A meta-analysis of all randomized trials comparing 5-FU with 5-FU modulate plus LV demonstrated that in patients with advanced colorectal cancer the response rate (RR) was significantly increased by using the 5-FU plus LV modulation, even though survival was not increased [6]. Several trials evaluated the modulation of 5-FU versus surgery alone in adjuvant treatment of colon cancer [7–10]. The results from these trials demonstrated significant improvements in diseasefree survival (DFS) and overall survival (OS) for the chemotherapy regimen. Among the various combinations of LV and 5-FU, in the USA the Mayo Clinic regimen for 6 months appears

Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 9, 2015

F. Di Costanzo1*, A. Sobrero2, S. Gasperoni1, L. Dogliotti3, L. Frassineti4, A. Falcone5, R. Lionetto6, P. Bruzzi6, G. Luppi1, L. Gallo2, P. Conte5, A. Comandone3, D. Turci4, M. Marzola1, U. Folco2, E. Pfanner5, M. Mestriner3, C. Boni1, C. Galli5, M. Tonato1 & R. Rosso2 On behalf of the Italian National Intergroup of Adjuvant Chemotherapy in Colon Cancer (INTACC)

1366 conducted every 6 months for the first 2 years and then every year for a total of 5 years. During follow-up, colonoscopy or barium enema was recommended every year during the first 5 years. Chest X-ray and hepatic ultrasound or computed tomography scan were performed every 6 months for the first 2 years, then every year until the fifth year. Treatment following relapse was decided by the investigators. Follow-up was continued after 5 years without formal protocol requirement.

Study end points and statistical methods

This trial was started in March 1992 and the accrual was completed in February 1995 after 1703 patients had been enrolled. Five cooperative groups took part in this multicenter trial: Gruppo Oncologico Italiano di Ricerca Clinica (GOIRC), Gruppo Cooperativo Istituto Nazionale per la Ricerca sul Cancro (IST-Genova), Gruppo Oncologico del Nord-Ovest (GONO), Gruppo Oncologico Piemontese Tumori Apparato Digerente (GOPTAD) and Istituto Oncologico Romagnolo (IOR). All patients were required to have histological proof of intraperitoneal colon carcinoma, complete resection of the primary tumor without gross or microscopic evidence of residual disease, age ≥18 years, stage B2–3 and C according to modified Astler and Coller (1987) classification [12], Eastern Cooperative Oncology Group (ECOG) performance status (PS) <1, normal hepatic and renal function, and informed consent. Colon cancer was defined as any lesion of the large bowel that did not require the opening of the pelvic peritoneum to define the distal extent of the tumor and/or 12 cm from pectinate line. The inferior margin of the tumor had to be above the peritoneal reflection.

INTACC aimed to recruit 725 patients per arm, which would give a >80% chance that the study could detect a 20% proportional difference in 5-year overall mortality between the two regimens (e.g. 60% versus 67%), based on a reported 3.5-year OS of 71% [3]. The final analysis was set when a total of 570 events had been observed. The primary end point was OS. Secondary end points included DFS and toxicity. Statistical analyses were based on the intention-to-treat principle. An interim analysis was planned and executed 2 years after recruiting the last patient to monitor study progression. A cut-off date of follow-up was set for 30 June 2000 to minimize the effect of potential bias in the reporting of events. Kaplan–Meier estimates were used to describe the time course of events in the treatment groups and the log-rank test was used for comparisons. Cox regression analysis was used to assess the relationship of the study end points to treatment after adjusting for a set of independent clinical variables, including ECOG performance status, histological grading, TNM (tumor–node–metastasis) stage of disease, gender and site of disease. Age, in 10-year intervals, was used as a stratification factor. The final models were obtained by means of a backward procedure based on the likelihood ratio test (P <0.05 for inclusion and ≥0.10 for elimination). The presence of significant interactions between treatment and prognostic factors was assessed by including the appropriate interaction terms in the model. Estimates were derived from a multivariate Cox’s model, with all prognostic factors included as covariates. From the total population of 1703 patients, 28 with missing information for one or more of the above factors were excluded. The P-value for heterogeneity of the hazard ratios was estimated by including interaction factors in the model. The proportional hazard assumptions were graphically checked with the log [–log (survival function)]. SPSS 9.0 statistical package was used for the statistical procedures. For all tests, a two-sided P-value of <0.05 was regarded as significant.

Randomization and treatment

Results

Patients and methods Eligibility criteria

Eligible patients were centrally randomized within 60 days after surgery at three coordinating data centers in Genova for IST-Genova, GONO and GOPTAD, and in Parma for GOIRC and Forlì for IOR. Randomization lists were generated using permuted blocks of variable length in random sequence so that treatments were balanced within each center. Clinical center was the only stratification factor. The two randomization arms were: arm A, 5-FU [450 mg/m2 intravenous (i.v.) bolus for 5 days, every 28 days] or arm B, 5-FU (370 mg/m2 i.v. bolus for 5 days, every 28 days) plus 6-S-LV (100 mg/m2 i.v. bolus on days 1–5, every 28 days), for a total of six cycles. In both arms, patients received LEVA (150 mg p.o. over 3 days, every 14 days). Toxicity was defined according to WHO criteria. 5-FU dose modifications were required for hematological and gastrointestinal toxicity. For gastrointestinal toxicities (grade 2/3) a dose reduction of 25% was foreseen. Dose reductions of LV and LEVA were not allowed.

Follow-up Patient data were collected on standardized trial forms and periodically sent to the Information Management Center. Investigators were required to notify the trial office of any serious unexpected adverse event. Before each course of therapy, patients underwent a physical examination and baseline hematological studies, including red and white blood cell count, renal function tests and serum carcinoembryonic antigen (CEA) measurements. These tests were

Patient characteristics Among the 1703 patients enrolled, 844 (49%) were randomized to arm A and 859 (51%) to arm B. The characteristics of the study population by treatment arm are shown in Table 1. Overall, patient and tumor characteristics were well balanced between the two arms. Fifty-eight patients (3%) were ineligible: 15 patients (0.8%) had unknown histology, four patients (0.2%) had PS >2, 29 patients (1.7%) because the interval from surgery to randomization was >60 days and 17 patients (0.9%) had colon cancer (Dukes’ B1) [13].

Treatment information As shown in Table 2, chemotherapy was completed according to protocol in 447 patients (53%) in arm A and in 413 patients (48%) in arm B, while treatment was modified in 254 patients (30%) in arm A and in 269 patients (31%) in arm B. The treatment was interrupted in 111 patients (13%) in the control arm and in 140 patients (16%) in the experimental arm. Sixteen (2%) and 18 patients (2%) did not start the treatment in the control arm and in the experimental arm, respectively. The median total dose of

Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 9, 2015

to be an effective treatment and has been used extensively as adjuvant therapy [7]. However, in the Mayo Clinic regimen, stomatitis (36%), diarrhea (44%) and leukopenia (14%) grade 3/4 according to the World Health Organization (WHO) were frequent adverse events. In Europe, the most used 5-day schedule of 5-FU plus LV was the regimen derived from the Machover schedule [8]. With this schedule the most common toxicities of grade 3/4 disease were: nausea and vomiting 4%, stomatitis 11% and diarrhea 8% [11]. Another important clinical issue that some trials have evaluated in the adjuvant setting, was the duration of treatment (12 versus 6 months). In 1992 the Italian National Intergroup of Adjuvant Chemotherapy in Colon Cancer (INTACC) designed a randomized trial for patients with curatively resected stage II and III colon cancer to evaluate whether the efficacy of the combination 5-FU plus LEVA for 6 months could be improved with the addition of LV.

1367 Table 1. Patient characteristics and clinical outcomes in the two treatment arms Arm Aa

Arm Ba

5-FU (450 mg/m2)

5-FU (370 mg/m2) + LV (100 mg/m2)

844

859

Median

63

63

Range

26–84

27–82

Characteristics

Randomized patients Age (years)

0

743 (88)

1

761 (88.5)

99 (11.7)

96 (11.1)

Right colon

243 (28.7)

272 (31.6)

Left colon

196 (23.2)

199 (23.1)

Sigmoid

384 (45.4)

371 (43.1)

Multiple

12 (1.42)

12 (1.4)

Unknown

9 (1.06)

Site (%)

5 (0.58)

Stage (%) Dukes’ B2–3

388 (46)

425 (49.4)

Dukes’ C ≤4 p.o. nodes

332 (39.3)

305 (35.5)

Dukes’ C >4 p.o. nodes

94 (11.1)

98 (11.5)

Dukes’ C, unknown nodes

23 (2.7)

21 (2.4)

1

90 (10.6)

105 (12.2)

2

563 (66.7)

563 (65.5)

3

120 (14.2)

113 (13.1)

WHO grade (%)

Unknown

71 (8.4)

78 (9)

73

74

Outcome Follow-up duration, months (median) Alive Total (%)

548 (64.9)

584 (67.9)

Disease-free

482

515

With recurrence Deaths (%)

66

69

296 (35.1)

275 (32)

232 (78.4)

214 (77.8)

Cause of death (%) Tumor Toxicity Other

6 (2.2) 30 (10)

1 (0.4) 30 (10.9)

Unknown

28 (9.4)

30 (10.9)

Recurrence

304 (36)

284 (33.1)

Data are given as median and relative range or as number (and per cent) of patients. P >0.05 for all comparisons. a Both arms were given levamisole 150 mg p.o. every 2 weeks. 5-FU, 5-fluorouracil; LV, leucovorin; p.o., per os.

5-FU delivered was 13.5 g in arm A and 10.8 g in arm B, with a duration of treatment of 24 weeks in both arms. As expected, the dose intensity (DI) of 5-FU administered for all cycles was significantly higher in arm A (535; 47–686 mg/m2/week) than in arm B

(437; 2–1050 mg/m2/week). The median number of courses was six (range zero to six): the proportion of patients who completed all scheduled courses was 84% in the control arm and 81% in the experimental arm.

Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 9, 2015

ECOG performance status (%)

1368

Toxicity

Clinical outcomes At 6.4 years [95% confidence interval (CI) 6.3–6.5] of median follow-up, a total of 571 of 1703 study patients had died. A similar number of deaths was seen in the two arms [296 (35%) versus 275 (32%)] (Table 1). Distribution of deaths by cause showed slightly more toxic deaths in arm A [5 (2.2%) versus 1 (0.4%)]. The most frequent causes of death were severe gastrointestinal toxicity (diarrhea, vomiting) associated with severe neutropenia, septic fever and other systemic compliances. The number of recurrences was similar in arm A (304, 36%) and arm B (284, 33%). The most common site of recurrence was the liver: 23% in both arms. The risk of a second primary tumor was similar in both arms (7% versus 8%). The most frequent second tumors were breast cancer (arm A versus arm B: eight versus two patients), colon cancer (arm A versus arm B: three versus six patients), gastric cancer (arm A versus arm B: two versus four patients) and myeloma (arm A versus arm B: two versus one patient). By Kaplan–Meier analysis, 5-year OS was not dissimilar between arms A (68%) and B (71%) (log-rank test, not significant). DFS at 5 years was 58% in arm A and 60% in arm B, respectively (Figure 1). In a subanalysis of both end points between treatments in patients stratified by disease stage (Dukes’ B2–3 versus C), no remarkable differences were seen. Multivariate Cox regression analysis did not show a significant association between treatment and either OS or DFS (Figure 2). Treatment B was associated with a nonsignificant reduction of 5% in the relative risk of death (hazard ratio 0.95; 95% CI 0.80–1.13; P = 0.57) and of 3% in the relative risk of relapse, second tumor or death (hazard ratio 0.97; 95% CI 0.83–1.13; P = 0.66). No significant interaction was seen between treatment and age, sex, the number of positive lymph nodes and tumor location, indicating the lack of any subgroup-specific treatment effect.

Discussion This trial did not show a statistically significant difference between 5-FU plus LV versus 5-FU alone in colon cancer patients treated for 6 months, considering that LEVA was administered with the same schedule in both treatment arms. However, this trial was not designed to evidence, nor was the statistical sample sufficient to

Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 9, 2015

Table 2 reports the incidence and types of toxicity associated with the two chemotherapy regimens. There were significant differences in toxicity between the two arms: severe gastrointestinal toxicity occurred more frequently in arm B (P <0.0000) while leukopenia grade 3/4 was more frequent in arm A (3% versus 0.8%; P = 0.02), respectively. Doses were reduced due to leukopenia in 106 patients (41%) in arm A and in 53 patients (18%) in arm B, due to diarrhea in 41 patients (16%) in arm A and in 87 patients (29%) in arm B, due to mucositis in 26 patients (10%) in arm A and 64 patients (21%) in arm B. Hepatic toxicity grade ≥3 was found in one patient in both arms. Two patients experienced multifocal leukoencephalopathy after the second and third cycle with 5-FU plus LEVA [14].

evidence, significant statistical differences from the addition of LV to 5-FU. Both tested regimens were associated with a satisfactory clinical outcome in terms of DFS (58% versus 60%) and OS (68% versus 71%), rates that are comparable with results from the literature. Although 5-FU plus LV is currently accepted as ‘standard’ treatment for colon cancer, the enigma concerning the value of LEVA in combination with 5-FU, or the value of 5-FU alone, remains unresolved. Some hypotheses can be proposed to explain the reasons for the equal activity of both regimens in this trial. Numerous investigators have suggested that the DI of 5-FU may be important in determining the activity of adjuvant chemotherapy in colon cancer [15]. Zalcberg et al. [16] evaluating the DI of 5-FU in adjuvant treatment, showed that the benefit associated with the use of 5-FU and LEVA in Dukes’ C colon cancer is directly related to the planned total dose of 5-FU administered. This benefit was greater in trials where the largest dose of 5-FU was combined with LEVA compared with other 5-FU regimens. The conclusion of Zalcberg et al.’s retrospective analysis suggests two hypotheses: first, the planned dosage of 5-FU is directly related to the results, when combined with LEVA; and secondly, the contribution of LEVA may not have had significant impact [16]. According to the Zalcberg hypothesis, the results of our trial can depend on the planned dose of 5-FU in arm A compared with that in arm B. It is possible that the effect of LV modulation was not evident due to counterbalancing by the 5-FU dose reduction in the experimental arm. A recent trial evaluated an adequate dose of 5-FU versus 5-FU plus LEVA and showed that 5-FU alone was as active as in combination with LEVA in prolonging survival in stage III colon cancer patients [17]. Recently, Tepper et al. [18] reported a large trial in patients with high-risk rectal cancer treated with adjuvant radiotherapy combined with 5-FU alone, 5-FU plus LEVA, 5-FU plus LV, and 5-FU plus LEVA and LV. No statistically significant advantage to any of the treatment regimens was found and data suggest that 5-FU is the only really effective agent in the adjuvant therapy [18]. In our trial, 5-FU was combined with LV in the experimental arm using the Machover schedule (the standard treatment in Europe), and we administered LEVA in both arms because at the onset of the trial it was not ethical to treat patients without 5-FU plus LEVA. In both arms, we administered chemotherapy for 6 months with the objective to increase the compliance of patients and to reduce the costs. Recent trials evaluating the duration of treatment (6 versus 12 months) have reported that 6 months treatment with 5-FU plus LV did not appear to be inferior to 12 months of treatment [10]. Haller et al. [19] (INT0089) demonstrated that 6 months of 5-FU plus LV administered by weekly and monthly schedule is equal to 5-FU/LEVA for 12 months, and supported that 6 months of 5-FU plus LV is a better treatment and the value of LEVA is irrelevant [19]. Wolmark et al.’s (NSABP-C04) [20] trial compared six cycles of 5-FU plus LV (Roswell Park schedule) to standard 5-FU plus LEVA (Moertel’s schedule) and double modulation of 5-FU plus LV and LEVA. The results did not reveal significant differences among the regimens. In 1992, in the INTACC trials, we used high doses of LV, but the optimal dose of LV is still under question. In particular, in an attempt to solve the question of LV dose and the

1369 Table 2. Treatment characteristics, dose delivery data and toxic effects Arm Aa

Arm Ba

5-FU (450 mg/m2)

5-FU (370 mg/m2) + LV (100 mg/m2)

Completed

447 (53.0)

413 (48.1)

Dose and timing modified

254 (30.1)

269 (31.3)

202 (24)

221 (25.7)

Characteristics

Chemotherapy (%)

Due to toxicity

65 (7.6)

65 (7.4)

Stopped

111 (13.1)

140 (16.3)

Due to death

5 (0.5)

4 (0.4)

Due to toxicity

41 (4.8)

62 (7.2)

Due to refusal

26 (3)

41 (4.7)

Other

42 (4.9)

35 (4.0)

16 (1.9)

18 (2.09)

Total 5-FU dose delivered (g)

13.5 (0.46–22.4)

10.8 (0.47–16.6)

Chemotherapy duration (weeks) (six cycles)

24 (4–51)

24 (4–38)

% of planned dose delivered

95

94.5

Never started 5-FU dose intensity (range)

2

Dose intensity, mg/m /week for all cycles

535 (47–686)

437 (2–1050)

Hematological toxicity (%) Total

133 (16.2)

68 (80.1); P <0.0000

24 (2.9)

7 (8.33); P <0.02

≥Grade 3 Leukopenia Grade 1–2

107 (12.7)

61 (7.1)

≥Grade 3

24 (2.8)

6 (0.7)

Thrombocytopenia Grade 1–2 ≥Grade 3

2 (0.1)

–

–

1 (0.1)

Anemia Grade 1–2 ≥Grade 3

1 (0.1) –

– –

Gastrointestinal toxicity (%) Total ≥Grade 3

553 (68.3)

649 (78.9); P <0.0000

70 (8.6)

152 (18.5); P <0.0000

Mucositis Grade 1–2

274 (33.0)

≥Grade 3

29 (3.5)

350 (41.5) 77 (9.1)

Diarrhea Grade 1–2

244 (29.3)

≥Grade 3

42 (5.1)

330 (39.2) 81 (9.6)

Nausea and vomiting Grade 1–2

354 (42.6)

≥Grade 3

16 (1.9)

411 (48.8) 27 (3.2)

Categorical data are expressed as number (and per cent) of patients, continuous variables as median (and relative range). a Both arms were given levamisole 150 mg p.o. every 2 weeks.

Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 9, 2015

Other

1370

role of LEVA, the Quick and Simple and Reliable (QUASAR) group designed a factorial (2 × 2) large-scale trial where the patients were randomly assigned either high-dose (175 mg fixed dose) or low-dose (25 mg fixed dose) LV, and either LEVA or placebo combined with 5-FU (370 mg/m2 for 5 days). The randomized comparison in the QUASAR group suggests that there are no survival differences between high- and low-dose LV or between LEVA and placebo [21]. Moreover, in a non-randomized trial, the QUASAR group evaluated the comparison between weekly and 4-weekly treatment schedules. The weekly regimen is less toxic than the 4-weekly regime, but there is no apparent difference in recurrence rates or OS [22]. In our trial, using the monthly schedule with LV plus 5-FU plus LEVA, the overall gastrointestinal toxicity, in particular diarrhea, stomatitis and vomiting, was statistically greater (32% versus 79%, arm A versus arm B; P <0.0000). Six patients died due to severe toxicity in the 5-FU plus LEVA arm, while one toxic death was related to the LV-combination treatment arm. In conclusion, our trial fails to demonstrate that the addition of LV to 5-FU plus LEVA is superior to 5-FU plus LEVA, both administered for 6 months. At this time, 5-FU plus LV is considered the standard adjuvant chemotherapy for colon cancer, although the new results that have emerged from advanced colon cancer treatment with CPT-11, oxaliplatin and oral fluoropyrimidine, justify new trials to increase and improve these results.

Acknowledgements We would like to thank Maureen Luxton for her help in the preparation and translation of this manuscript. Supported by clinical grants from the Associazione Italiana per la Ricerca sul Cancro (AIRC) and Consiglio Nazionale Ricerche,

Progetto Finalizzato Applicazioni Cliniche della Ricerca Oncologica (CNR-ACRO). Presented in part at the Congress of the American Society of Clinical Oncology (ASCO), Atlanta, GA, USA, 1999. INTACC investigators are listed below. GOIRC: V. Silingardi, A. Frassoldati, S. Zironi, Divisione di Oncologia Medica Ospedale Policlinico, Modena; C. Modonesi, II Divisione di Medicina Generale, Sez. Aggregata di Oncologia Medica, Arcispedale S. Anna, Ferrara; E. Corgna, Divisione di Oncologia Medica, Policlinico Monteluce, Perugia; L. Giustini, L. Acito, Servizio di Oncologia, Ospedale Civile, Fermo; A. Contu, N. Olmeo, G. Baldino, A. Pazzola, Oncologia Medica ASL N.1 Ospedale Civile, Sassari; A. Lalli, F. Fabbri, Servizio di Oncologia Presidio Ospedaliero, Giulianova; S. Barni, A. Ardizzoia, U. O. di Radioterapia e Oncologia, Monza; C. Rodinò, Servizio di Oncologia Ospedale Civile, Piacenza; V. Pieroni, G. Giacomini, U. Oncologia ASL 5, Jesi; G. De Signoribus, F. Giorgi, Servizio di Oncologia, Ospedale Civile, San Benedetto del Tronto; F. Pucci, F. Leopardi, R. Camisa, Oncologia Ospedale Maggiore, Parma; M. Lombardo, P. Di Stefano, Dipartimento di Oncologia ed Ematologia, Unità Operativa di Chemioimmunoterapia, Pescara; F. Recchia, S. De Filippis, Divisione di Medicina Oncologia, Ospedale Civile, Avezzano; F. Bassan, A. RosaBian, Servizio Autonomo di Oncologia Medica, Presidio Ospedaliero, Tiene; F. Cavicchi, M. Alessio, M. Sisani, Istituto di Medicina Interna e Scienze Oncologiche Policlinico Monteluce, Perugia; L. Di Lullo, Medicina Generale, Ambulatorio di Oncologia, Ospedale Civile, Isernia. Genova: C. Pucetti, M. Giusto, Servizio Oncologia Medica, Osp. Civile, Belluno; S. Monfardini, D. Crivellari, U.O. Oncologia Medica, Azienda Ospedale, Aviano; G. Accarpio, M. Mari, E. Paganini, G. Becchi, Sez. Aggregata Oncologia Pad. 5/3, A. O. Villa Scassi, Genova; S. Mammoliti, U.O. Oncologia Medica I, Istituto Nazionale per la

Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 9, 2015

Figure 1. Kaplan–Meier estimates of overall survival (left) and disease-free survival (right) according to treatment arm.

Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 9, 2015

Figure 2. Effect of treatment assignment on overall survival across strata defined by various prognostic factors. PS, performance status; CL, confidence limit.

1371

1372

References 1. Coleman MP, Esteve J, Damiecki P et al. Trends in cancer incidence and mortality. IARC Scientific Publication No. 121. Lyon, France: IARC 1993. 2. Cohen AM, Shank B, Friedman MA. Colorectal cancer. In De Vita VT, Hellman S, Rosemberg SA (eds): Cancer Principles and Practice Oncology, 3rd edition. Philadelphia, PA: Lippincot 1989; 859–964. 3. Moertel CG, Fleming TR, Mcdonald JS et al. Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. N Engl J Med 1990; 322: 352–358. 4. NIH consensus conference. Adjuvant therapy for patients with colon and rectal cancer. JAMA 1990; 264: 1444–1450. 5. Sobrero A, Aschele C, Bertino J. Fluorouracil in colorectal cancer. A tale of two drugs: implications for biochemical modulation. J Clin Oncol 1997; 15: 368–381. 6. 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.

7. O’Connell MJ, Mailliard JA, Khan MJ et al. Controlled trial of fluorouracil and low-dose leucovorin given for 6 months as postoperative adjuvant therapy for colon cancer. J Clin Oncol 1997; 5: 246–250. 8. International Multicentre Pooled Analysis of Colon Cancer Trials (IMPACT). Efficacy of adjuvant fluorouracil and folinic acid in colon cancer. Lancet 1995; 345: 939–944. 9. Francini G, Petrioli R, Lorenzini L et al. Folinic acid and 5-fluorouracil as adjuvant chemotherapy in colon cancer. Gastroenterology 1994; 106: 899–906. 10. O’Connel MJ, Laurie JA, Kahn M et al. Prospectively randomized trial of postoperative adjuvant chemotherapy in patients with high-risk colon cancer. J Clin Oncol 1998; 16: 295–300. 11. Machover D. A comprehensive review of 5-fluorouracil and leucovorin in patients with metastatic colorectal carcinoma. Cancer 1997; 80: 1179– 1187. 12. Beahrs OH, Myers MH, American Joint Committee on Cancer (eds): Manual for Staging of Cancer, 3rd edition. Philadelphia, PA: Lippincot 1987. 13. Di Costanzo F. For the Intergruppo Nazionale Terapia Adiuvante Carcinoma Colon (INTACC) INTACC 01: 5-fluorouracil (5-FU) vs levamisole (LEVA) vs 5-FU + 6-S-leucovorin (6-S-LV) + LEVA: an Italian Intergroup Study of Adjuvant Therapy for Colon Cancer. Proc Am Soc Clin Oncol 1999; 18: 266 (Abstr). 14. Luppi G, Zoboli A, Barbieri F et al. On behalf of the INTACC. Multifocal leukoencephalopathy associated with 5-fluorouracil and levamisole adjuvant therapy for colon cancer. A report of two cases and review of the literature. Ann Oncol 1996; 7: 412–415. 15. Wolmark N, Bryant J, Smith R. Adjuvant 5-fluorouracil and leucovorin with or without interferon alfa-2a in colon carcinoma: National Surgical Adjuvant Breast and Bowel Project Protocol C-05. J Natl Cancer Inst 1998; 90: 1810–1816. 16. Zalcberg JR, Siderov J, Simes J. On behalf of the Australasian Gastrointestinal Trials Group. The role of 5-fluorouracil dose in the adjuvant therapy of colorectal cancer. Ann Oncol 1996; 7: 41–46. 17. Cascinu S, Catalano V, Latini L et al. A randomized trial of adjuvant therapy of stage III colon cancer: levamisole and 5-fluorouracil versus 5-fluorouracil alone. Proc Am Soc Clin Oncol 1999; 18: 240a (Abstr). 18.. Tepper JE, O’Connel MJ, Petroni GR et al. Adjuvant postoperative fluorouracil-modulated chemotherapy combined with pelvic radiation therapy for rectal cancer: initial results of intergroup 0114. J Clin Oncol 1997; 15: 2030–2039. 19. Haller DG, Catalano PJ, Macdonald JS, Mayer RJ. Fluorouracil, leucovorin, and levamisole adjuvant therapy for colon cancer: five years final report of INT-0089. Proc Am Soc Clin Oncol 1998; 17: 256 (Abstr). 20. Wolmark N, Rockette H, Mamounas EP et al. Clinical trial to assess the relative efficacy of fluorouracil and leucovorin, fluorouracil and levamisole, and fluorouracil, leucovorin and levamisole in patients with Dukes’ B and C carcinoma of the colon: results from National Surgical Adjuvant Breast and Bowel Project C-04. J Clin Oncol 1999; 17: 3553– 3559. 21. QUASAR Collaborative Group. Comparison of fluorouracil with additional levamisole, higher-dose folinic acid, or both, as adjuvant chemotherapy for colorectal cancer. Lancet 2000; 355: 1588–1596. 22. Kerr DJ, Gray R, McConkey C, Barnwell J. Adjuvant chemotherapy with 5-fluorouracil, L-folinic acid and levamisole for patients with colorectal cancer: non-randomized comparison of weekly versus four-weekly schedules—less pain, same gain. QUASAR Colorectal Cancer Study Group. Ann Oncol 2000; 11: 947–995.

Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 9, 2015

Ricerca sul Cancro, Genova; A. Lavarello, Servizio Oncologia, USL 18-Ospedale, Genova; F. Brema, G. Pastorino, Servizio Oncologia, Ospedale S. Paolo, Savona; G. Bottero, V. Fusco, Divisione Oncologia, Ospedale S. Antonio, Alessandria; G. Orsi, D. Cocchis, Ospedale Civile, Fossano; A. Guglielmi, U.O. Oncologia Medica, Ospedale S. Martino, Genova. GONO: P. Pronzato, Divisione Oncologia Medica, La Spezia; G. Barsanti, M. Battistoni, Divisione Oncologia Medica, Lucca; P. Grezzi, M. Rinaldini, Divisione Oncologia Medica, Arezzo; R. Taviani, Divisione Oncologia Medica, Empoli; M. Aglietta, C. Volta, Divisione Clinica Medica, Novara; R. Algeri, Divisione Oncologia Medica, Grosseto; P. De Gennaro, Divisione Chirurgia, Bibbiena; I. Spinelli, Divisione Oncologia Medica, Massa-Carrara; M. Filidei, Divisione Medicina, Pontedera; G. Arcabasso, Divisione Medicina, Cecina; A. Carli, Divisione Chirurgia, Siena. GOPTAD: R. Antonacci, Oncologia Medica Università di Torino, Osp. San Luigi, Orbassano; C. Bumma, R. Berardo, Oncologia Medica, Osp. S. Giovanni Antica Sede, Torino; A. Boglione, Oncologia Medica, Osp. Gradenigo, Torino; A. Di Napoli, Medicina Interna, Osp. Molinette sede San Vito, Torino; M. Merlano, G. Grecchi, Oncologia Medica, Osp. S. Croce & Carle, Cuneo; G. Porcile, M. Destefanis, Oncologia Medica, Osp. San Lazzaro, Alba; O. Alabiso, R. Buosi, Oncologia Medica Università del Piemonte Orientale, Osp. Maggiore della Carità, Novara; M. Botta, B. Castagneto, Oncologia Medica, Osp. Santo Spirito, Casale Monferrato; V. M. Zampaglione, P. Sozzi, Medicina A, Osp. Degli Infermi, Biella; V. Sidoti, Oncologia Medica, Osp. Agnelli, Pinerolo; S. Di Prima, Oncologia Medica, Osp. Santa Croce, Moncalieri. IOR: D. Amadori, O. Nanni, P. Rosetti, Dipartimento di Oncologia Ospedale Pierantoni, Forlì; M. Marangolo, F. Zumaglini, Dipartimento di Oncologia Ospedale Civile, Ravenna; D. Tassinari, E. Pasquini, Divisione di Oncologia Medica Ospedale Infermi, Rimini; F. Artioli, M. Cesari, Unità Operativa di Oncologia Medica Ospedale Ramazzini Carpi; A. Gambi, L. Amaducci, Divisione di Oncologia Ospedale Civile per gli Infermi, Faenza; P. Alessandrini, G. Catalano, Unità Operativa di Oncologia Medica Azienda Ospedaliera S. Salvatore, Pesaro.