Post-Marketing Safety Study of Ramucirumab Plus FOLFIRI: Analysis of Age and Initial Dose of Irinotecan in Patients with Metastatic Colorectal Cancer

This was a single-arm, prospective, multicenter, non-interventional, observational study in patients with mCRC who were administered ramucirumab plus FOLFIRI under routine clinical practice. The study was conducted between December 2016 and April 2020. Patients were enrolled across 73 medical institutions in Japan.

Patients diagnosed with unresectable, advanced, or recurrent mCRC and treated for the first time with ramucirumab plus FOLFIRI under routine clinical practice were eligible for study inclusion. The study design has been previously described [11]. Briefly, on day 1 of each 2-week cycle, patients received intravenous ramucirumab 8 mg/kg followed by FOLFIRI (intravenous irinotecan 150 or 180 mg/m², intravenous levofolinate 200 mg/m², and fluorouracil 400 mg/m² given as an intravenous bolus injection followed by 2400 mg/m² given as a continuous infusion over 46 h). Dosage could be tailored to each patient’s medical condition. The duration of observation was 12 months from the start of ramucirumab administration. If the use of ramucirumab was discontinued, the last day of the observation period was to be 30 days after the last administration or the day the post-treatment was started, whichever was earlier.

Data were reported by investigators via an electronic data capture (EDC) system using Classic Rave, versions 2016.2.0 and 2018.2.4 (Medidata Solutions, Inc., NY, USA). Where possible, patient registration occurred within 14 days from administration of ramucirumab. Data collected via the EDC system include enrollment data, baseline patient characteristics, prior treatment for CRC, concomitant drugs and therapies, performance status, laboratory test values, adverse events (AEs), laboratory test values related to AEs, survival (investigated 12 months after initiation of ramucirumab), and continuation or discontinuation of ramucirumab. AEs were defined as any unfavorable or unintended disease or sign (including an abnormal laboratory test value) in a patient who was administered ramucirumab plus FOLFIRI, with or without a causal relationship to ramucirumab. AEs were graded using Common Terminology Criteria for Adverse Events (CTCAE), version 4.02. Serious AEs (SAEs) were events that resulted in death, disability, or a congenital anomaly/birth defect, were life threatening or required initial or prolonged hospitalization, or other medically serious conditions. Notable AEs included arterial thromboembolic events, venous thromboembolic events (VTEs), infusion-related reactions, gastrointestinal perforation, bleeding/hemorrhagic events, congestive cardiac failure, wound healing complications, fistula, posterior reversible encephalopathy syndrome, hypertension, proteinuria, liver failure/injury, interstitial lung disease, neutropenia, febrile neutropenia, and leukopenia. Notable AEs were summarized using consolidated terms comprising one or more Medical Dictionary for Regulatory Activities (MedDRA) preferred terms.

Statistical analyses were based on estimation, as this study was performed under routine clinical practice. The study population was stratified by age (≥ 75 years vs < 75 years, ≥ 70 years vs < 70 years, and ≥ 65 years vs < 65 years) and initial dose of irinotecan (≤ 150 mg/m² vs > 150 mg/m²). The primary analysis for age was set to ≥ 75 years versus < 75 years. The results of the statistical analyses were presented using mean, median, incidence proportions, and point estimates with associated confidence intervals (CIs). Categorical variables, including binary variables, were summarized using statistics such as frequency and incidence proportion. Continuous data were summarized using mean (standard deviation) or median (range). Survival status was estimated using the Kaplan–Meier method, including the 12-month survival rate and associated 95% CI. Statistical analyses were performed using SAS Version 9.4.

Of the 366 patients who enrolled in the study, four patients from the safety and effectiveness analysis population were excluded because of enrollment criteria violation (three patients with missing signatures and one patient with a prior history of ramucirumab treatment) and seven patients were excluded from the effectiveness analysis population because of first-line treatment with ramucirumab plus FOLFIRI. Overall, 362 patients were included in the safety analysis population and 355 were included in the efficacy analysis population.

Patient demographics and baseline clinical characteristics by age and dose of initial irinotecan treatment are outlined in Table 1. Regardless of age, more patients were male (52.6% aged ≥ 75 years and 55.7% aged < 75 years). The incidence of RAS mutation was slightly higher in patients aged ≥ 75 years (n = 35, 61.4%) than in those aged < 75 years (n = 167, 54.8%). Older patients aged ≥ 75 years were more likely to have the ascending colon as a primary tumor site (n = 19, 33.3%) than patients aged < 75 years (n = 60, 19.7%). There were no substantial differences in baseline patient characteristics between patients aged < 65 and ≥ 65 years or < 70 and ≥ 70 years (Supplementary Table S1, see Electronic Supplementary Material [ESM]). Patient demographics and baseline clinical characteristics were generally similar by dose of initial irinotecan treatment. There were no differences in Eastern Cooperative Oncology Group (ECOG) performance status by initial irinotecan doses of ≤ 150 mg/m² and > 150 mg/m², and more than half of patients in both populations had RAS gene mutations (irinotecan ≤ 150 mg/m²: n = 152, 58.2%; irinotecan > 150 mg/m²: n = 46, 52.3%).

The median (range) dose intensity of ramucirumab was 2.8 (1.1–4.2) versus 3.0 (0.9–4.4) mg/kg/week (≥ 75 vs < 75 years; Table 2). Patients aged ≥ 75 years received fewer median cycles of ramucirumab plus FOLFIRI than those aged < 75 years (ramucirumab: 5 vs 8 cycles; FOLFIRI: 6 vs 8 cycles). The median dose intensity of irinotecan and fluorouracil was 46.0 versus 55.3 mg/m²/week and 776.0 versus 1006.2 mg/m²/week in patients aged ≥ 75 years and < 75 years, respectively. The dose of ramucirumab and FOLFIRI in combination with levofolinate, irinotecan, and fluorouracil by ages < 65, ≥ 65, < 70, and ≥ 70 years is shown in Supplementary Tables S2–S5 (see ESM).

The frequency of any grade AEs and grade ≥ 3 AEs by age (≥ 75 years vs < 75 years) was 86.0% (n = 49) versus 83.9% (n = 256) and 56.1% (n = 32) versus 50.2% (n = 153), respectively, indicating no substantial differences between the two age populations. SAEs were reported in 26.3% (n = 15) of patients aged ≥ 75 years and in 18.7% (n = 57) of those aged < 75 years (Table 3).

The most commonly occurring any grade notable AEs in both age groups included neutropenia (≥ 75 years: 40.4%, n = 23; < 75 years: 41.6%, n = 127), hypertension (≥ 75 years: 15.8%, n = 9; < 75 years: 16.4%, n = 50), and proteinuria (≥ 75 years: 17.5%, n = 10; < 75 years: 17.7%, n = 54). The incidence of any grade VTEs was higher in patients aged ≥ 75 years than in those aged < 75 years (7.0% vs 1.3%). There were no obvious differences in grade ≥ 3 notable AEs between patients aged ≥ 75 years and those aged < 75 years. In addition, there were no substantial differences in the frequency of any grade AEs, grade ≥ 3 AEs, or notable AEs in patients aged < 65 versus those aged ≥ 65 years or in patients aged < 70 versus those aged ≥ 70 years (Supplementary Table S6, see ESM).

In terms of effectiveness, the 12-month survival rate following ramucirumab plus FOLFIRI treatment was lower in older patients: 47.2% (95% CI 32.9–61.4) for patients aged ≥ 75 years and 61.1% (95% CI 55.2–66.9) for those aged < 75 years (Fig. 1). The 12-month survival rates were not substantially different in patients aged < 65 versus those aged ≥ 65 years or in patients aged < 70 versus those aged ≥ 70 years (Supplementary Figure S1, see ESM).

The frequency of grade ≥ 3 AEs was slightly lower in patients receiving > 150 mg/m² of irinotecan (42.1%, n = 37) than in those receiving ≤ 150 mg/m² of irinotecan (53.6%, n = 140). SAEs were reported in 17.1% (n = 15) of patients receiving > 150 mg/m² of irinotecan and in 19.9% (n = 52) of those receiving ≤ 150 mg/m² of irinotecan (Table 3).

In terms of notable AEs, the frequency of grade ≥ 3 liver failure injury was higher in patients receiving an initial irinotecan dose of > 150 mg/m² than in those receiving ≤ 150 mg/m² (9.1%, n = 8 vs 2.3%, n = 6). There were no substantial differences in the frequency of any grade diarrhea by initial irinotecan dose (> 150 mg/m² vs ≤ 150 mg/m²: n = 13, 14.8% vs n = 44, 16.9%); however, the frequency of grade ≥ 3 diarrhea was higher in patients who received an initial irinotecan dose of > 150 mg/m² (n = 4, 4.6%) than in those who received ≤ 150 mg/m² (n = 5, 1.9%). Overall, hematological toxicities and grade ≥ 3 febrile neutropenia were similar between irinotecan doses of > 150 mg/m² and ≤ 150 mg/m².

There was no substantial difference in the 12-month survival rate between initial irinotecan doses of ≤ 150 mg/m² (61.0%, 95% CI 54.7–67.4) and > 150 mg/m² (51.4%, 95% CI 40.0–62.9; Fig. 1).