A multicentre, randomized pilot trial comparing vascular access strategies for early stage breast cancer patients receiving non-trastuzumab containing chemotherapy.

All vascular access strategies foradministering chemotherapy in early stage breast cancer (EBC) are associated with risks and benefits. As the most effective type of access is unknown a feasibility trial, prior to conducting a large pragmatic trial, was undertaken. The trial methodology utilized broad eligibility criteria and the integrated consent model incorporating oral consent. EBC patients receiving non-trastuzumab-containing chemotherapy were randomized to peripheral access or central line insertion. The a priori definition of feasibility was: > 25% of patients approached agreed to randomisation and > 25% of physicians approached patients. Secondary outcomes included rates of line-associated complications. Of 159 patients approached, 150 (94.3%) agreed to randomisation, 77 (51.3%) were randomized to peripheral and 73 (48.7%) to central access. 6/26 (23.1%) of medical oncologists approached patients. Rates of complications per chemotherapy cycles in the peripheral vs central access groups with risk difference (RD) (95% CI) were: thrombotic events requiring anticoagulation [1 (0.3%) vs. 3 (1.0%), RD - 0.7(- 1.9,0.5)], line infections [0 (0%) vs. 1 (0.3%), RD - 0.3(- 0.9,0.3)], phlebitis [2 (0.6%) vs. 0 (0%), RD 0.3(- 0.3,0.8)], and tissue infiltrations [4 (1.1%) vs. 1 (0.3%), RD 0.8(- 0.4,2.1)]. Overall, 8.0% (6/75) and 7.7% (5/65) of patients had at least one of these complications in the peripheral and central access arms respectively [RD - 0.9(- 9.4,7.6)]. The study was terminated early due to slow accrual. While meeting its a priori feasibility criteria for patient engagement, the slow accrual means that conducting a large pragmatic trial would require overcoming the barriers to physician recruitment. NCT02688998.