Introduction

A regimen of infusional 5-fluorouracil/leucovorin (5-FU/LV) plus irinotecan (FOLFIRI) followed by infusional 5-FU/LV plus oxaliplatin (FOLFOX) or the reverse sequence has, until recently, been the standard treatment strategy for metastatic colorectal cancer (mCRC) patients. However, in a ramdomized phase III trial conducted by the Groupe Coopérateur Multidisciplinaire en Oncologie (GERCOR), FOLFIRI resulted in a response rate of 4% and median progression-free survival (PFS) of 2.5 months in patients previously treated with FOLFOX [1]: these efficacy results were unsatisfactory. Recently, bevacizumab (a humanized monoclonal antibody that inhibits vascular endothelial growth factor [VEGF], a key mediator of angiogenesis) in combination with oxaliplatin- or irinotecan-containing regimens has become the standard of care for untreated patients with mCRC [2–5]. In the Eastern Cooperative Oncology Group (ECOG) E3200 study, bevacizumab plus FOLFOX showed a higher response rate (22.7% vs. 8.6%), longer PFS (median, 7.3 vs. 4.7 months; hazard ratio = 0.61; P < .0001), and longer overall survival (OS) (median, 12.9 vs. 10.8 months; hazard ratio = 0.75; P = .0011) compared with FOLFOX alone in the patients previously treated with combination irinotecan and fluoropyrimidine chemotherapy [6]. However, the additive benefit of bevacizumab to FOLFIRI after failure of oxaliplatin-containing therapy remains unclear.

The aim of the present study was to evaluate the activity of bevacizumab plus FOLFIRI after failure of oxaliplatin and fluoropyrimidine in the clinical practice setting.

Patients and methods

After bevacizumab was approved in Japan in June 2007, we initiated an observational cohort study to evaluate the efficacy and safety of bevacizumab treatment combined with FOLFIRI or FOLFOX after failure of chemotherapy in bevacizumab-naïve mCRC patients. Patient registration was controlled centrally by the Tsukuba Cancer Clinical Trial Group (TCTG).

Patients were included in this study if they (1) were ≥20 years old, (2) had a pathologic diagnosis of colorectal adenocarcinoma, (3) either had (a) prior chemotherapy stopped due to disease progression or unacceptable toxicity or (b) relapsed during or within 6 months following the last administration of preoperative or postoperative chemotherapy, and (4) had then received bevacizumab plus FOLFIRI or bevacizumab plus FOLFOX starting between June 2007 and March 2008. Patients all had an ECOG performance status (PS) of ≤2 and adequate organ function. The exclusion criteria were: first-line chemotherapy including bevacizumab, bevacizumab plus FOLFIRI administered after failure of irinotecan-containing therapy, or bevacizumab plus FOLFOX administered after failure of oxaliplatin-containing therapy.

This study was conducted according to the Japanese Ethical Guidelines for Epidemiological Research [7] and was approved by the Ethics Committee of each participating hospital. This study was registered with the University Hospital Medical Information Network (No. UMIN 000001547).

Chemotherapy

Bevacizumab was administered at a dose of 5 or 10 mg/kg every 2 weeks in combination with chemotherapy, based on physician choice. FOLFIRI consisted of infusional 5-FU/LV combined with irinotecan every 2 weeks, while FOLFOX consisted of infusional 5-FU/LV combined with oxaliplatin every 2 weeks. These chemotherapy regimens were continued until disease progression, unacceptable toxicity, or patient refusal. Dose reductions, delays, and discontinuation of chemotherapy were allowed at the discretion of the treating physician(s).

Assessments

Baseline data included patient background, prior chemotherapy regimens (fluoropyrimidine-, oxaliplatin-, irinotecan-containing, or other), and initial doses of bevacizumab and irinotecan or oxaliplatin. In addition, whether an epidermal growth factor receptor (EGFR) inhibitor was administered after discontinuation of bevacizumab combination chemotherapy was noted. Patient responses to chemotherapy, severe adverse events (grade ≥ 3), bevacizumab discontinuations due to bevacizumab-related adverse events, date of recorded progression, and death were collected every 6 months until either death, loss to follow-up, or study closure on August 31, 2010. Investigators assessed tumor response and disease progression according to the Response Evaluation Criteria In Solid Tumors (RECIST, version 1.0) every 2 or 3 months. Adverse events were assessed according to the National Cancer Institute Common Terminology Criteria for Adverse Events (version 3.0) at every hospital visit. Reported adverse events included bevacizumab-related toxicities and any other adverse events of grade ≥3.

Statistical analysis

OS was calculated as the period from the first day of bevacizumab treatment to death from any cause or to the date of last follow-up (censored). PFS was calculated as the period from the first day of bevacizumab treatment to either the date of recorded tumor progression, death from any cause, or last follow-up without disease progression (censored). OS and PFS were analyzed using the Kaplan–Meier method, with a confidence interval (CI) of 95%. Statistical analysis was performed using SPSS software, version 18.0 (IBM Japan, Tokyo).

Results

Patients

A total of 195 patients who failed with bevacizumab-naïve chemotherapy were enrolled in this observational cohort study, and 160 of these were evaluated in the present study (Fig. 1). Of the evaluated patients, 115 patients received bevacizumab plus FOLFIRI after failure of oxaliplatin and fluoropyrimidine (FOLFIRI+BV after OX/FU group), and 45 received bevacizumab plus FOLFOX after failure of irinotecan and fluoropyrimidine (FOLFOX+BV after IRI/FU group). The median follow-up time was 18.4 months (range, 0.6–36.8 months).

Fig. 1
figure 1

Patient flow

Full size image

The patient characteristics of each group were similar (Table 1). Most patients had an ECOG PS of ≤1. Seven patients (6.1%) in the FOLFIRI+BV after OX/FU group had previously received FOLFOX in the adjuvant setting, and two patients (4.4%) in the FOLFOX+BV after IRI/FU group had previously received FOLFIRI in the adjuvant setting. A bevacizumab dose of 10 mg/kg every 2 weeks was used in 9 patients (7.8%) in the FOLFIRI+BV after OX/FU group and in 7 patients (16%) in the FOLFOX+BV after IRI/FU group. After bevacizumab combination chemotherapy, an EGFR inhibitor was administered to 39 patients (34%) in the FOLFIRI+BV after OX/FU group and 14 patients (31%) in the FOLFOX+BV after IRI/FU group. The KRAS status of these patients is not known.

Table 1 Patient characteristics
Full size table

The major reasons for discontinuation of bevacizumab combination chemotherapy were disease progression (72% in the FOLFIRI+BV after OX/FU group and 73% in the FOLFOX+BV after IRI/FU group) and toxicity (20% in the FOLFIRI+BV after OX/FU group and 22% in the FOLFOX+BV after IRI/FU group).

Efficacy

The number of reported PFS events was 103 (90%) in the FOLFIRI+BV after OX/FU group and 43 (96%) in the FOLFOX+BV after IRI/FU group. Median PFS was 8.3 months (95% CI, 6.7–9.9) for the FOLFIRI+BV after OX/FU group and 7.8 months (95% CI, 5.8–9.7) for the FOLFOX+BV after IRI/FU group (Fig. 2a). The number of reported deaths was 83 (72%) and 34 (76%), respectively. Median OS was 21.6 months (95% CI, 17.6–25.6) and 16.5 months (95% CI, 11.8–21.2), respectively (Fig. 2b).

Fig. 2
figure 2

Progression-free survival (a) and overall survival (b). FOLFIRI+BV after OX/FU bevacizumab plus FOLFIRI after failure of oxaliplatin and fluoropyrimidine (solid line), FOLFOX+BV after IRI/FU bevacizumab plus FOLFOX after failure of irinotecan and fluoropyrimidine (dotted line)

Full size image

The 7 patients in the FOLFIRI+BV after OX/FU group who had failed adjuvant FOLFOX experienced a median PFS of 12.5 months (95% CI, 3.8–29.0) and a median OS of 28.6 months (95% CI, 14.5–not reached).

The number of patients with measurable lesions was 103 (90%) in the FOLFIRI+BV after OX/FU group and 41 (91%) in the FOLFOX+BV after IRI/FU group. No patient in either group experienced a complete response. The overall response rate was 25% in the FOLFIRI+BV after OX/FU group and 29% in the FOLFOX+BV after IRI/FU group.

Safety

Grade ≥3 adverse events are listed in Table 2. The overall incidence of grade ≥3 bevacizumab-related adverse events was 11% in the FOLFIRI+BV after OX/FU group and 13% in the FOLFOX+BV after IRI/FU group. The most common events were hypertension (5.0%) and bleeding (3.8%). No gastrointestinal perforation was reported. Reported bevacizumab discontinuations due to bevacizumab-related adverse events were 15 (13%) in the FOLFIRI+BV after OX/FU group and 4 (9%) in the FOLFOX+BV after IRI/FU group. The major reason for discontinuation was bleeding episodes that occurred in six (grade 3) and five (grade 2). In all cases, bleeding episodes ceased after discontinuation of bevacizumab.

Table 2 Grade ≥3 adverse events
Full size table

Grade ≥3 adverse events not related to bevacizumab occurred in 55 patients (48%) in the FOLFIRI+BV after OX/FU group and in 26 patients (58%) in the FOLFOX+BV after IRI/FU group. Neutropenia was the main toxicity in both groups. The grade ≥3 non-hematological toxicity rate was <10%, except for peripheral neuropathy (16%) in the FOLFOX+BV after IRI/FU group.

Four patients (3.5%) in the FOLFIRI+BV after OX/FU group and one (2.2%) in the FOLFOX+BV after IRI/FU group died within 30 days of the last treatment. These included three treatment-related deaths (2 due to interstitial pneumonitis and one due to pulmonary embolism).

Discussion

The benefit of adding bevacizumab to FOLFOX in previously treated patients with mCRC was shown in the E3200 study [6]. However, the efficacy of FOLFIRI+BV after OX/FU has remained unclear. In the present study, the efficacy results for FOLFOX+BV after IRI/FU (29% response rate, 7.8 months of median PFS) were comparable to those of FOLFOX+BV in the E3200 study (23% response rate, 7.3 months of median PFS), and FOLFIRI+BV after OX/FU also showed similar favorable outcomes (25% response rate, 8.3 months of median PFS). These results appear to be better than those of the second-line FOLFOX without bevacizumab or FOLFIRI without bevacizumab, which were reported in the GERCOR trial [1]. Taken together, the present study demonstrates that the addition of bevacizumab may provide a substantial benefit to second-line chemotherapy with FOLFIRI after failure of an oxaliplatin-containing regimen.

In general, bevacizumab is dosed at 5 mg/kg every 2 weeks when combined with first-line chemotherapy for mCRC; however, it was administered at 10 mg/kg every 2 weeks in the E3200 study, because it was used in the second-line setting, which is associated with a greater tumor burden. The optimal dose of bevacizumab (5 or 10 mg/kg/every 2 weeks) for the second-line chemotherapy setting for mCRC remains controversial. In the present study, the majority of patients were treated with bevacizumab 5 mg/kg/every 2 weeks, and our results suggest that this dose is sufficient, even within the second-line setting.

This study has some biases because it was an observational cohort study, so the present results must be confirmed in a randomized controlled study. However, most patients with mCRC are currently treated with bevacizumab plus FOLFOX or bevacizumab plus FOLFIRI as the standard first-line chemotherapy; thus, it is challenging to conduct a prospective trial to assess the benefit of bevacizumab added to second-line chemotherapy in bevacizumab-naïve patients. Thus, the results of the present study seem quite relevant. In particular, the present results in patients who received FOLFIRI+BV after OX/FU can be applied to the selection of the optimal second-line treatment regimen when adjuvant chemotherapy has failed. The standard adjuvant chemotherapy strategy for Stage III colon cancer is oxaliplatin-containing chemotherapy (FOLFOX4 in the MOSAIC trial, bolus 5-fluorouracil/leucovorin plus oxaliplatin [FLOX] in the NSABP C-07 trial, and capecitabine plus oxaliplatin [XELOX] in the XELOXA trial) [8–10]. Based on the results of the present study, if recurrence occurs during or within 6 months after adjuvant chemotherapy, bevacizumab plus FOLFIRI may be recommended.

In a randomized phase III trial, the anti-EGFR antibody panitumumab combined with FOLFIRI as a second-line treatment resulted in a median PFS of 5.9 months and a median OS of 14.5 months in patients harboring wild-type KRAS [11]. Patients who received panitumumab plus FOLFIRI had received prior oxaliplatin (67%) and bevacizumab (18%). These results are similar to those observed in patients who received FOLFIRI+BV after OX/FU in the present study. It has been suggested that patients with wild-type KRAS mCRC can elect to receive panitumumab plus FOLFIRI after oxaliplatin-containing chemotherapy. However, it is well known that the magnitude of the survival benefit of bevacizumab appears to be larger within first-line therapy than in second- or third-line therapy, and is not influenced by tumor KRAS status. In contrast, panitumumab or cetuximab after failure of 5-FU, oxaliplatin, and irinotecan has been shown to improve survival in the third-line setting [12, 13]. Thus, these findings suggest that bevacizumab plus FOLFIRI could be recommended rather than anti-EGFR antibodies to bevacizumab-naïve patients as earlier lines of treatment.

FOLFIRI+BV after OX/FU was well tolerated. In previous clinical trials and large multinational observational studies, the most common grade ≥3 bevacizumab-related adverse events were hypertension (5–20%), bleeding (3–4%), and thromboembolism (3–19%) in patients receiving either first- or second-line bevacizumab combination chemotherapy [6, 14–17], and treatment discontinuation due to any adverse event ranged from 23 to 30%. The safety profiles and toxicity severity observed in these previous reports are comparable to the observations in patients who received FOLFIRI+BV after OX/FU in the present study.

Conclusion

FOLFIRI+BV after OX/FU showed an adequately acceptable toxicity and efficacy similar to that for FOLFOX+BV after IRI/FU in clinical practice.