This study investigated the magnitude and etiology of neuromuscular fatigue and muscle damage induced by eccentric cycling compared with conventional concentric cycling in patients with breast cancer. After a gradual familiarization protocol for eccentric cycling, nine patients with early-stage breast cancer performed three cycling sessions in eccentric or concentric mode. The eccentric cycling session (ECC) was compared with concentric cycling sessions matched for power output (CON power ; 80% of concentric peak power output, 95 ± 23 W) or oxygen uptake ( ; 10 ± 2 mL·min·kg -1 ). Preexercise to postexercise changes (30-s through 10-min recovery) in knee extensor maximal voluntary contraction force (MVC), voluntary activation, and quadriceps potentiated twitch force ( Qtw ) were quantified to determine global, central, and peripheral fatigue, respectively. Creatine kinase and lactate dehydrogenase activities were measured in the plasma before and 24 h after exercise as markers of muscle damage. Compared with CON power (-11% ± 9%) and (-5% ± 5%), the ECC session resulted in a greater decrease in MVC (-25% ± 12%) postexercise ( P < 0.001). Voluntary activation decreased only in ECC (-9% ± 6% postexercise, P < 0.001). The decrease in Qtw was similar postexercise between ECC and CON power (-39% ± 21% and -40% ± 16%, P > 0.99) but lower in ( P < 0.001). The CON power session resulted in twofold greater compared with the ECC and sessions ( P < 0.001). No change in creatine kinase or lactate dehydrogenase activity was reported from preexercise to 24 h postexercise. The ECC session induced greater neuromuscular fatigue compared with the concentric cycling sessions without generating severe muscle damage. ECC is a promising exercise modality for counteracting neuromuscular maladaptation in patients with breast cancer.