Colorectal cancer is one of the most common malignancies in the western world, with over 300,000 new cases diagnosed in Europe and the United States each year. Surgery is the principal treatment for operable patients, but for patients who have relapsed following surgery or who present with metastatic disease, the prognosis is poor. Although systemic treatment is possible, the effect is mainly palliative. For almost 40 years, fluorouracil(Drug information on fluorouracil) (5-FU) has been the mainstay of the treatment of patients with advanced colorectal cancer.
Over the past 15 years, several attempts have been made to increase the efficacy of 5-FU, either through modulation of its activity with compounds such as leucovorin, methotrexate(Drug information on methotrexate), or interferon-alpha, or through techniques that increase tumor exposure to the drug by means of protracted continuous infusion. Although these methods have increased the efficacy of 5-FU by a factor of two to three in terms of response rate, survival of patients with advanced colorectal cancer remains relatively disappointing, rarely exceeding a median of 12 months.
Treatment options for these patients remained limited until recently, when the topoisomerase I inhibitor irinotecan(Drug information on irinotecan) (CPT-11, Camptosar) and the diaminocyclohexane (DACH) platinum compound oxaliplatin(Drug information on oxaliplatin) (EDITOR'S NOTE: roxaliplatin is investigational in the United States and available as Eloxatin in France) were developed and approved for use in advanced colorectal carcinoma.
Irinotecan is a water-soluble semisynthetic compound derived from the antineoplastic agent camptothecin, a natural alkaloid of the Chinese tree Camptotheca acuminata. It is a selective inhibitor of the DNA enzyme topoisomerase I. In preclinical studies, irinotecan and its active metabolite SN38 demonstrated significant antitumor activity against a variety of tumor cell lines in vitro and also showed activity in in vivo models, including tumors (such as colorectal cancer) that express the p-glycoprotein-mediated, multidrug-resistant phenotype.
Irinotecan has demonstrated activity as a single agent in several phase I and phase II trials conducted in Europe, Japan, and the United States. Initial phase I studies of irinotecan began in Japan in 1986, and the most common dose-limiting toxicities were found to be hematologic (neutropenia) and gastrointestinal (delayed diarrhea). Early in its development, irinotecan was tested as a short infusion (30 to 90 min) using three different schedules: (1) 100 mg/m2/d over 3 days, repeated every week; (2) 100 to 125 mg/m2/d every week; and (3) 350 mg/m2 every 3 weeks.
The most widely used schedule of irinotecan in European phase I studies was a single dose administered every 3 weeks. These studies led to the recommendation that a dose of 350 mg/m2 every 3 weeks be evaluated in clinical phase II trials. Delayed diarrhea was countered by subsequent incorporation of high-dose loperamide(Drug information on loperamide).
In a multicenter phase II trial in 166 patients with previously treated colorectal cancer, Rothenberg et al administered irinotecan at 125 mg/m2/wk to the first 64 patients and at 100 mg/m2 to the subsequent 102 patients. Irinotecan was administered as a 90-minute infusion for 4 consecutive weeks followed by 2 weeks’ rest. There were no significant differences in grade 3/4 toxicities between those receiving 125 mg/m2 and those receiving 100 mg/m2, except for grade 3/4 emesis (22% vs 2%; P < .001. Compared with younger patients, elderly patients (> 65 years old) were twice as likely (38.6% vs 18.8%, P < .008) to develop grade 3/4 diarrhea.
Objective responses to irinotecan were observed in 18 patients (1 complete response and 17 partial responses) for an overall response rate of 10.8% (95% confidence interval (CI) = 6.1%-15.6%). In addition, 67 patients (40.4%) had stable disease. The treatment response rates were 14.1% (9/64) for the 125-mg/m2 arm and 8.8% (9/102) for the 100-mg/m2 arm. This trial suggested a trend toward a higher response rate without substantially greater toxicity for the 125-mg/m2 dose, while the overall median survival was 9.9 months (range: 0.3 to 38.8 months, Table 1).
Multicenter Phase III Trials: Two important randomized multicenter phase III studies of irinotecan as a single agent were performed during its development in Europe. The first, conducted by Rougier et al, compared irinotecan with continuous-infusion 5-FU in 267 advanced colorectal cancer patients who either did not respond or whose disease had progressed after first-line treatment with 5-FU. Patients were randomly assigned to receive a 90-minute infusion of irinotecan at 300 to 350 mg/m2 every 3 weeks or a continuous infusion of 5-FU according to three different schedules (Table 2). Both treatments were given until the development of disease progression or unacceptable toxicity.
As second-line therapy, irinotecan improved survival and median time to disease progression compared to continuous-infusion 5-FU, with relative outcomes of 10.8 vs 8.5 months (P = .035) and 4.2 vs 2.9 months (P = .035), respectively (Figure 1 and Figure 2). Both treatments were well tolerated except for vomiting and diarrhea, and patients were able to maintain good quality of life (Figure 3) and tumor control. Because irinotecan has proven to significantly improve progression-free and overall survival compared to continuous-infusion 5-FU after failure of therapy with standard 5-FU, it should be considered one of the reference drugs for second-line treatment of advanced colorectal cancer.
The second trial, conducted by Cunningham et al in 279 advanced colorectal cancer patients, compared irinotecan plus best supportive care vs best supportive care alone after 5-FU failure. A 90-minute infusion of irinotecan at 300 to 350 mg/m2 every 3 weeks plus best supportive care was administered to 189 patients, while another 90 patients received best supportive care alone (Figure 4). At a median follow-up of 13 months, overall survival was significantly better in the irinotecan arm (P = .0001) (Figure 5). The 1-year survival reported for the irinotecan arm was 36.2%, compared to 13.8% for the best supportive care arm, thus showing a clear advantage for salvage chemotherapy with irinotecan.
A multivariate analysis adjusted to prognostic factors significantly favored irinotecan when assessed in terms of survival (P = .001), survival without performance status deterioration (P = .0001), survival without a weight loss of more than 5% (P = .018), and pain-free survival (P = .003) (Figure 6). Despite the presence of treatment-induced diarrhea, significant differences in the quality-of-life analysis were also found to favor irinotecan. This study showed that despite the side effects, patients with metastatic colorectal cancer in whom 5-FU therapy has failed have a longer survival with fewer tumor-related symptoms and better quality of life when they are treated with irinotecan, as compared to best supportive care alone.
Four Different Schedules: A recent randomized phase II multicenter trial compared four different schedules of irinotecan as second-line therapy in advanced colorectal cancer patients. Patients were randomized to arm A (350 mg/m2 every 3 weeks), arm B (125 mg/m2, weekly, for 4 weeks, in 6-week cycles), arm C (250 mg/m2 every 2 weeks), or arm D (10 mg/m2/d by continuous infusion for 14 days every 3 weeks).
All four regimens were well tolerated and demonstrated efficacy. Arm C (every 2 weeks) and arm A (every 3 weeks) showed similar results for survival and tumor control while showing better efficacy compared to the weekly schedule (arm B). Continuous-infusion irinotecan (arm D) did not show the expected efficacy or safety (producing diarrhea, nausea, and vomiting), although it was associated with a low incidence of grade 3/4 neutropenia. Therefore, the new standard schedule to be adopted for further development will consist of irinotecan administered every 2 or 3 weeks.
Several combination regimens containing irinotecan have been tested in Europe and the United States in recent years. The superiority of irinotecan over best supportive care and other second-line chemotherapy in colorectal cancer patients suggested that it would be of interest to combine irinotecan with other active agents such as 5-FU, oxaliplatin, or mitomycin(Drug information on mitomycin) (mitomycin-C [Mutamycin]).
Irinotecan and 5-FU or 5-FU/Leucovorin: A phase I trial and pharmacokinetic analysis performed between 1995 and 1997 at the Pitié-Salpêtrière Hospital in 41 patients with solid tumors (29 advanced colorectal cancer patients), demonstrated the feasibility and safety of irinotecan in combination with 5-FU (Table 3).[7,8] Irinotecan was administered on day 0 followed by 5-FU on days 1 to 5 in cycle 1 and on day 6 after 5-FU in cycle 2. The aim of the trial was to determine the relationship between the sequence of irinotecan administration and toxicity when combined with 5-FU.
The maximum tolerated dose of irinotecan was established at 350 mg/m2. For further phase II trials, the investigators recommended that irinotecan be given at 200 mg/m2 in combination with 5-FU at 375 mg/m2 and leucovorin at 20 mg/m2 on days 1 and 5 (Mayo Clinic regimen). No difference in the clearance of irinotecan or its active metabolite SN38 was observed, but a reduction in the catabolism of 5-FU was demonstrated, suggesting a potential benefit if irinotecan is administered before 5-FU.[7,8]
In a phase I study conducted in Spain by Aranda et al, irinotecan was administered on day 1 at 150 to 250 mg/m2 followed by a 14-day continuous-infusion of 5-FU at 250 mg/m2/d and 1 week of rest in a variety of solid tumors. Three partial responses (gastric, colon, and rectum) and two minor responses (colon) were observed.[9,10]
Barone et al conducted a multicenter study in Italy in 33 advanced colorectal cancer patients who were treated with irinotecan at 350 mg/m2, over a 90-minute infusion on day 1 every 3 weeks, plus 5-FU and leucovorin for 5 consecutive days according to the Mayo Clinic schedule as first-line treatment. This combination was well tolerated despite moderate neutropenia and diarrhea and achieved an overall response rate of 31% with a median progression-free survival of 7.2 months and an overall survival of 16 months.
Rothenberg et al conducted another multicenter phase II trial in which 71 previously untreated patients received weekly irinotecan at 100 mg/m2 for 4 weeks every 6 weeks and 5-FU/leucovorin over 5 days (according to the Mayo Clinic schedule) every 4 weeks. The main toxicities were diarrhea and neutropenia (without severe consequences). The overall response rate with this regimen was 32% (95% CI: 21.5%-43.3%), and the median survival was 17.8 months.
Rougier et al recently published the results of a phase I/II trial in which irinotecan was administered in combination with FUFOL (5-FU and leucovorin [folinic acid]) according to the de Gramont schedule in 55 advanced colorectal cancer patients who had previously been treated with 5-FU (Table 4). This combination was tolerated well, with no cumulative toxicity and produced a response rate of 22%. The maximum tolerated dose of irinotecan was 300 mg/m2 dose level; thus, the recommended dose for phase II trials was a 90-minute infusion of irinotecan at 180 mg/m2 on day 1, and biweekly leucovorin/5-FU on days 1 and 2 every 2 weeks. A pharmacokinetic analysis performed in 21 patients showed that the maximum concentration of SN38 appeared 30 to 60 minutes after the administration of irinotecan (Table 4).
These results have been confirmed at our institution in 39 patients with advanced colorectal cancer, using the recommended dose of irinotecan (180 mg/m2 on day 1) and biweekly leucovorin/5-FU on days 1 and 2. Eight patients achieved an objective response (overall response rate = 20.5%), and 12 demonstrated disease stabilization or minor responses (tumor control = 49%).[14,15] Of eight patients who underwent hepatic resection, seven achieved a complete response. The median duration of response and the median survival were 14 and 11 months, respectively. Neutropenia was the most serious side effect (affecting 29% of patients in 2% of cycles), with four patients developing febrile neutropenia and grade 3 diarrhea.[14,15]
De Gramont and colleagues performed a phase II study to evaluate the FOLFIRI combination (irinotecan, 180 mg/m2 on day 1, plus a simplified biweekly leucovorin/5-FU regimen, with the daily 5-FU bolus omitted) in heavily pretreated colorectal cancer patients (Figure 7). The treatment was well tolerated, and the main toxicities were nausea and vomiting. Of 34 evaluable patients, 2 achieved a partial response (overall response rate: 6%), and disease stabilization was observed in 20, for a tumor control rate of 65% (Table 5).
Irinotecan and Mitomycin: A phase I/II study was performed at our institution to assess the feasibility of irinotecan in combination with mitomycin, another active compound in the treatment of gastrointestinal tract tumors, including advanced colorectal cancer. Between January 1996 and June 1997, the study accrued 26 heavily pretreated patients with advanced gastrointestinal cancer (22 with colorectal cancer). Mitomycin was administered as an IV bolus, and irinotecan as a 30-minute infusion. Four dose levels of irinotecan/mitomycin were evaluated: 300/8 mg/m2, 325/8 mg/m2, 350/8 mg/m2, and 325/10 mg/m2 administered on day 1 every 21 days (Table 6).
The dose-limiting toxicity was neutropenia, but it was of short duration and associated with rapid recovery. Cumulative thrombocytopenia occurred at the highest dose level as did one case of delayed hemolytic-uremic syndrome (both of which were due to mitomycin). Other toxicities included diarrhea and four cases of febrile neutropenia but no toxic deaths (Table 6).
Objective responses were reported at all dose levels, for an overall response rate of 28% (which is higher than expected for this heavily pretreated population and suggests a possible synergism between the two drugs) (Table 6). The recommended doses for further studies were 325 mg/m2 of irinotecan and 8 mg/m2 of mitomycin.
Irinotecan and Oxaliplatin: Irinotecan in combination with oxaliplatin has also shown activity in 5-FU-refractory patients. Cvitkovic et al recently performed three phase I/II studies in which the combination was administered following two different schedules:
Groups A and B received irinotecan, 150 to 250 mg/m2, plus oxaliplatin, 85 to 110 mg/m2, every 3 weeks.
Group C received irinotecan, 100 to 200 mg/m2, plus oxaliplatin, 85 mg/m2, every 2 weeks.
Of 34 patients, 15 responded (for an overall response rate of 44%), and toxicity was acceptable (Table 7). The recommended doses were irinotecan, 200 mg/m2, and oxaliplatin, 85 mg/m2, every 3 weeks (group A) or irinotecan, 175 mg/m2, and oxaliplatin, 85 mg/m2, every 2 weeks (group C).
Wasserman et al published a pharmacokinetic analysis of two independent phase I studies of the combination of irinotecan and oxaliplatin in patients with gastrointestinal tumors. Oxaliplatin was first administered as a 120-minute infusion followed by irinotecan as a 30-minute infusion, every 3 weeks. Pharmacokinetic analyses were performed during the first and second cycle. The maximum tolerated dose was reached at level 3, but two patients with Gilbert’s syndrome developed severe neutropenia, thrombocytopenia, and diarrhea.
Of 24 evaluable patients, 7 achieved a partial response (overall response rate: 29%), and 9 (38%) experienced disease stabilization. The pharmacokinetic parameters of the combined regimen were similar to those of the single agents.
Based on previous data, a phase I/II study and pharmacokinetic analysis of four drugs in combination (irinotecan, oxaliplatin, and biweekly 5-FU/leucovorin) is ongoing at our center in patients with advanced colorectal cancer. To date, 34 patients have been enrolled to receive oxaliplatin on day 1 (in a 2- to 4-hour infusion), irinotecan on day 2 (30-minute infusion), and leucovorin/5-FU (de Gramont schedule) in a fixed dose on days 2 and 3, every 2 weeks. Pharmacokinetic samples are drawn during cycles 1 and 2. In the second cycle, irinotecan is given on day 1 and oxaliplatin on day 2 (Table 8).
Five dose-levels are being evaluated: level 1 = oxaliplatin, 65 mg/m2, and irinotecan, 150 mg/m2; level 2 = oxaliplatin, 75 mg/m2, and irinotecan, 165 mg/m2; level 3 = oxaliplatin, 85 mg/m2, and irinotecan, 165 mg/m2; level 4 = oxaliplatin, 85 mg/m2, and irinotecan, 180 mg/m2. Dose level 5 is ongoing: oxaliplatin, 100 mg/m2, and irinotecan, 180 mg/m2. To date, the maximum tolerated dose has not been reached but preliminary results show high antitumor activity with acceptable toxicity at all dose levels (Table 8).
Of 24 evaluable patients in this study, 2 achieved complete responses and 8 achieved partial responses, for an overall response rate of 41%; 6 of these patients underwent subsequent partial hepatectomy and have achieved a complete response (Table 8). Preliminary pharmacokinetic results have revealed that oxaliplatin clearance is significantly lower (27.87 vs 23.24 L/h, P = .0047) when irinotecan is administered on day 1 before oxaliplatin, as compared to the reverse schedule.
Irinotecan and Raltitrexed: In a recent phase II study conducted by Nobile et al, the combination of irinotecan and raltitrexed was evaluated in 14 patients with advanced colorectal cancer. Irinotecan was given at 300 to 350 mg/m2 on day 1, and raltitrexed at 3 mg/m2 on day 2, with the cycle repeated every 3 weeks. The main toxicities were grade 3/4 diarrhea in 14% of patients, grade 3/4 nausea/vomiting in 42% of patients, and asthenia in 35.7% of patients; the incidence of febrile neutropenia was rare (7%). Of 10 evaluable patients, 3 achieved a partial response, 1 a minor response, and 4 disease stabilization (overall response rate: 30%). This represents an encouraging level of tumor control for this combination (Table 9).
Phase III Trials of Irinotecan/FUFOL Combinations: Two important randomized trials of irinotecan and FUFOL in combination have been conducted in the United States and Europe. The first trial, performed by Saltz et al, compared weekly irinotecan plus FUFOL to FUFOL (Mayo Clinic regimen) or weekly irinotecan alone in patients with previously untreated metastatic colorectal cancer. This trial demonstrated that the combination achieves better results than either single-agent therapy. The overall response rates were 49%, 27%, and 29% for the combination, irinotecan, and FUFOL, respectively. Although more grade 3/4 diarrhea and vomiting were observed with the combination, more grade 4 neutropenia and mucositis were seen with 5-FU/leucovorin alone.
The second randomized phase III trial, conducted by Douillard et al, compared FUFOL (at high doses, either according to the AIO [Arbeitsgruppe Internistische Onkologie, of the German Cancer Society] or de Gramont regimen) with or without irinotecan. Included in the trial were 385 previously untreated metastatic colorectal cancer patients: 145 received irinotecan plus biweekly leucovorin/5-FU and 143 received FUFOL alone, while 54 and 43 patients received FUFOL according to the AIO regimen, in combination with irinotecan or FUFOL/AIO alone, respectively.
More toxicity was observed with the combination than with FUFOL alone, resulting in higher rates of grade 3/4 neutropenia (42% vs 11% of patients) and febrile neutropenia (5% vs 1% of patients), as shown in Table 10. In terms of nonhematologic grade 3/4 toxicities, patients who received the combination experienced more diarrhea than those who received FUFOL alone (22% vs 10% of patients). Of 169 patients evaluable for response in both arms, 4% achieved a complete response, and 37% a partial response, for an overall response rate of 41% for irinotecan plus 5-FU/leucovorin and a 23% overall response rate for the 5-FU/leucovorin combination (P < .001).
A substantial number of patients experienced disease stabilization in both groups (38% and 50%, respectively) (Table 10); however, the disease-free survival and median overall survival (Figure 8) was also significantly superior (16.8 vs 14 months, P = .031) in favor of the irinotecan/FUFOL combination, with good quality-of-life results (Figure 9).
A pooled analysis of these two phase III randomized trials, comparing irinotecan plus FUFOL vs FUFOL alone in advanced colorectal cancer patients who received no prior chemotherapy showed that the time to progression and overall survival were significantly higher for the irinotecan plus FUFOL combination in both studies, as well as in the combined analysis (time to progression: P < .001; overall survival: P < .009). In conclusion, the combination of irinotecan and FUFOL as first-line treatment significantly improved tumor control and overall survival compared to FUFOL alone. This combination may, therefore, be considered the new standard for first-line treatment of advanced colorectal cancer.