Noninferiority of Four vs Six Cycles of CHOP for Patients With Favorable Prognosis B-Cell Lymphoma: FLYER Trial

Get Permission

As reported in The Lancet by Viola Poeschel, MD, and colleagues, the phase III FLYER trial has shown noninferiority of progression-free survival with four vs six cycles of cyclophosphamide/doxorubicin/vincristine/prednisone (CHOP) in combination with six doses of rituximab in patients with aggressive B-cell lymphoma with a favorable prognosis.

Viola Poeschel, MD

Viola Poeschel, MD

Study Details

The open-label noninferiority trial included 588 patients from 138 sites in Denmark, Israel, Italy, Norway, and Germany. Patients had to be between age 18 and 60 and have stage I or II disease; normal serum lactate dehydrogenase concentration; Eastern Cooperative Oncology Group performance status of 0 or 1; and no bulky disease (maximal tumor diameter < 7.5 cm).  The primary endpoint was investigator-assessed progression-free survival at 3 years, with a selected noninferiority margin of -5.5%.

Patients were randomly assigned between December 2005 and October 2016 to receive four cycles of R-CHOP plus two doses of rituximab as monotherapy (n = 293) or six cycles of R-CHOP (n = 295).

CHOP consisted of 750 mg/m² of cyclophosphamide, 50 mg/m² of doxorubicin, and 1.4 mg/m² of vincristine (maximum total dose of 2 mg) on day 1 and prednisone (or prednisolone) at 100 mg on days 1 to 5 of 21-day cycles. Rituximab was given at 375 mg/m² on day 1 of each cycle for six cycles, with the final two doses in the four-cycle group given as monotherapy.

Progression-Free Survival

Median follow-up was 66 months. Three-year progression-free survival was 96% (95% confidence interval [CI] = 94%–99%) in the four-cycle group vs 94% (95% CI = 91%–97%) in the six-cycle group, thus satisfying the noninferiority criterion. On multivariate analysis including stage I vs II disease, extralymphatic vs lymphatic involvement, and presence vs absence of B symptoms, the hazard ratio for progression-free survival for the four-cycle vs six-cycle group was 0.9 (95% CI = 0.5–1.6, P = .810).  

At the end of therapy, 91% vs 92% of patients had complete response or unconfirmed complete response and 3% vs 4% had partial response. Three-year event-free survival was 89% (95% CI = 86%–93%) vs 89% (95% CI = 85%–92%). Three-year overall survival was 99% (95% CI = 98%–100%) vs 98% (95% CI = 96%–99%). On multivariate analysis including the same factors as above, the hazard ratio for overall survival for the 4-cycle vs 6-cycle group was 0.9 (95% CI = 0.4–1.9, P = .722).

In post hoc analysis, estimated 5-year progression-free survival was 94% vs 94%, estimated 5-year event-free survival was 87% vs 88%, and estimated 5-year overall survival was 97% vs 98%.


  • Four cycles of CHOP plus six doses of rituximab was noninferior to six R-CHOP cycles in 3-year progression-free survival.
  • At 3 years, progression-free survival was 96% vs 94% and overall survival was 99% vs 98%.

Safety Profiles

Totals of 294 hematologic and 1,036 nonhematologic adverse events of any grade were reported in the 4-cycle group vs 426 and 1,280 in the six-cycle group. The most common grade 3 or 4 hematologic toxicities in both dosing groups were leukocytopenia (80 vs 110 patients) and thrombocytopenia (5 vs 7 patients). Nonhematologic grade 3 or 4 adverse events occurred in 52 vs 71 patients. Infection of any grade occurred in 116 patients in the four-cycle group (with 22 being grade 3 or 4) vs 156 in the six-cycle group (with 23 being grade 3 or 4). Other than infection, the most common grade 3 or 4 nonhematologic adverse events were paresthesia (16 vs 14 patients) and nausea (6 vs 12 patients).

Serious adverse events occurred in 48 vs 45 patients. Cardiac events (atrial fibrillation, heart failure, or coronary artery disease) were reported during therapy and follow up in four vs three patients. Secondary neoplasms were observed in 18 vs 14 patients. Death related to secondary neoplasms occurred in three vs four patients. Two patients, both in the six-cycle group, died during study treatment, both due to pneumonia.

The investigators concluded, “In young patients with aggressive B-cell non-Hodgkin lymphoma and favorable prognosis, four cycles of R-CHOP is noninferior to six cycles of R-CHOP, with relevant reduction of toxic effects. Thus, chemotherapy can be reduced without compromising outcomes in this population.”

Dr. Poeschel, of Saarland University Medical School, Homburg, and Gerhard Held, MD, of Westpfalz-Klinikum, Kaiserslautern, Germany, are the corresponding authors for The Lancet article.

Disclosure: The study was funded by Deutsche Krebshilfe. For full disclosures of the study authors, visit