Pharmacokinetics, safety, and efficacy of subcutaneous versus intravenous rituximab plus chemotherapy as treatment for chronic lymphocytic leukaemia (SAWYER): a phase 1b, open-label, randomised controlled non-inferiority trial

Lancet Haematol. 2016 Mar;3(3):e128-38. doi: 10.1016/S2352-3026(16)00004-1.


Background: Part one of the two-part SAWYER study predicted that subcutaneous rituximab 1600 mg would achieve trough serum concentrations that were non-inferior to those achieved with intravenous rituximab 500 mg/m(2) in patients with chronic lymphocytic leukaemia. In part two of the study, we aimed to confirm the pharmacokinetic non-inferiority of subcutaneous rituximab, and investigate its safety and efficacy.

Methods: We did this phase 1b, open-label, randomised controlled non-inferiority study at 68 centres in 19 countries in Europe, North America, South America, and Australasia. Patients aged 18 years or older with untreated chronic lymphocytic leukaemia were randomly assigned, via an interactive voice-response system with a permuted block randomisation scheme (block size of ten), to receive subcutaneous rituximab 1600 mg or intravenous rituximab 500 mg/m(2) plus fludarabine and cyclophosphamide every 4 weeks for up to six cycles. In cycle one, all patients received intravenous rituximab 375 mg/m(2). Randomisation was stratified by Binet stage and fludarabine and cyclophosphamide administration route (oral vs intravenous). Study investigators and patients were not masked to group allocation, but allocation was concealed from the statistician, clinical scientist, and clinical pharmacologist. The primary endpoint was trough serum concentration at cycle five, with a non-inferiority margin of 0·8 for the adjusted geometric mean ratio of the subcutaneous to the intravenous dose. We did the primary analysis in patients in the intention-to-treat population with complete pharmacokinetic data (pharmacokinetic population). This trial is registered with, number NCT01292603, and is ongoing, although the treatment stage is now complete.

Findings: Between Aug 20, 2012, and June 17, 2013, we randomly assigned 176 patients to receive subcutaneous rituximab (n=88) or intravenous rituximab (n=88); 134 (76%) patients comprised the pharmacokinetic population. As of May 7, 2014, median follow-up was 13·9 months (IQR 11·9-16·0) for patients in the subcutaneous group and 14·1 months (11·6-16·5) for patients in the intravenous group. At cycle five, the geometric mean trough serum concentration in patients given subcutaneous rituximab was non-inferior to that in patients given intravenous rituximab (97·5 μg/mL vs 61·5 μg/mL), with an adjusted geometric mean ratio of 1·53 (90% CI 1·27-1·85). In the safety analysis, the proportion of patients reporting adverse events was similar between the subcutaneous and intravenous groups (all grades: 82 [96%] of 85 patients and 81 [91%] of 89 patients; serious adverse events: 25 [29%] and 29 [33%] patients; grade ≥3: 59 [69%] and 63 [71%] patients, respectively). The most common adverse event of grade 3 or higher was neutropenia (48 [56%] patients in the subcutaneous group and 46 [52%] patients in the intravenous group); the most common serious adverse event was febrile neutropenia (n=9 [11%] and n=4 [4%], respectively). We recorded administration-related reactions in 37 (44%) patients given subcutaneous rituximab and 40 (45%) patients given the intravenous dose, with differences between administration routes for injection-site erythema (n=10 [12%] and n=0, respectively) and nausea (n=2 [2%] and n=11 [12%], respectively). More patients reported local cutaneous reactions after subcutaneous rituximab (n=36 [42%]) than after intravenous rituximab (n=2 [2%]); most of these reactions were grade 1 or 2.

Interpretation: When combined with fludarabine and cyclophosphamide, subcutaneous rituximab 1600 mg achieved trough serum concentrations that were pharmacokinetically non-inferior to those achieved with intravenous rituximab 500 mg/m(2), with a similar safety and efficacy profile between the two groups. Treatment with subcutaneous rituximab should allow patients with chronic lymphocytic leukaemia to receive clinical benefit from the drug via a more convenient delivery method than the intravenous route, and might also be used in combination regimens with approved and emerging oral regimens.

Funding: F Hoffmann-La Roche.

Publication types

  • Clinical Trial, Phase I
  • Multicenter Study
  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Administration, Cutaneous
  • Administration, Intravenous
  • Adult
  • Aged
  • Antineoplastic Agents* / administration & dosage
  • Antineoplastic Agents* / adverse effects
  • Antineoplastic Agents* / pharmacokinetics
  • Antineoplastic Agents, Alkylating / administration & dosage
  • Antineoplastic Agents, Alkylating / adverse effects
  • Antineoplastic Agents, Alkylating / pharmacokinetics
  • Antineoplastic Combined Chemotherapy Protocols
  • Cyclophosphamide / administration & dosage
  • Cyclophosphamide / adverse effects
  • Cyclophosphamide / pharmacokinetics
  • Female
  • Humans
  • Leukemia, Lymphocytic, Chronic, B-Cell / drug therapy*
  • Male
  • Middle Aged
  • Rituximab* / administration & dosage
  • Rituximab* / adverse effects
  • Rituximab* / pharmacokinetics
  • Treatment Outcome
  • Vidarabine / administration & dosage
  • Vidarabine / adverse effects
  • Vidarabine / analogs & derivatives
  • Vidarabine / pharmacokinetics


  • Antineoplastic Agents
  • Antineoplastic Agents, Alkylating
  • Rituximab
  • Cyclophosphamide
  • Vidarabine
  • fludarabine

Associated data