Metabolic flexibility over peak power: Why some cyclists resist fatigue better than others

Endurance cycling performance is influenced by physiological factors including maximal oxygen uptake ([Formula: see text]O 2 max), sustainable [Formula: see text]O 2 , and gross efficiency (GE), all of which decline during prolonged exercise.. Durability is defined as the ability to maintain these physiological characteristics during exercise, despite progressive increases in peripheral fatigue. Yet, the underlying mechanism(s) for why these characteristics are better preserved in some individuals remains to be determined. Therefore, we aimed to explore physiological determinants of durability during unfatigued and fatigued cycling in individuals whose cycling durability differed. Eleven trained cyclists (n=3 females, 38±12 years, 174±11 cm, 76±12 kg, [Formula: see text]O 2 max 53.1±8 ml/kg/min, 11±5 hours/week) participated in three sessions: (1) a graded exercise test to determine maximal aerobic power (MAP), anaerobic threshold (AnT) and baseline cycling gross efficiency (GE), (2) a 40km fresh time trial (FRESH), and (3) a 40km time trial preceded by 4x8 mins at the power output associated with the AnT (PAnT) (FATIGUE). FRESH and FATIGUE were performed in a randomized order. Cyclists were retrospectively classified as HIGH durability (< 3% performance decrement between FRESH vs FATIGUE) or LOW durability ( >3% decrement). Group differences in physiological responses during both trials were evaluated. Quadriceps prolonged low-frequency force depression (PLFFD) was quantified from 10 Hz stimulation of the femoral nerve and near-infrared spectroscopy characterized muscle oxygen delivery and extraction. Maximal aerobic power was lower in HIGH (298±80 W) compared to LOW (388±56 W, p = 0.06, d=1.27). No significant group differences were seen in any other measured anthropometric or physiological characteristic including: PAnT (HIGH: 2.7±0.7 W/kg vs LOW: 3.0±0.4 W/kg, p=0.53, d=0.39) and GE at the AnT (HIGH: 19.4±1.5% vs LOW: 19.3±1.3%, p=0.87, d=0.10). During the trials, power output declined significantly less in HIGH (p=0.003,ηp2=0.69). A significant group x condition interaction was observed for power output relative to AnT (p=0.004, ηp2=0.67) such that during the trials, %AnT increased in LOW during FRESH and decreased in both HIGH and LOW during FATIGUE. %AnT was significantly reduced in LOW (from 101±5%AnT to 86±5%AnT, p=0.028). %AnT was maintained in HIGH across trials (p=0.24). PLFFD increased by ~60% following each trial, and was similar between trials (p=0.73) and groups (p=0.53). RER was significantly lower, and declined faster in HIGH (p=0.01, ηp2=0.54). GE declined throughout both trials (p=0.03), but was not significantly different between groups (p=0.74). Compared to FRESH, O 2 delivery was significantly lower in FATIGUE in LOW only (p=0.02). Higher durability is characterized by lower VO 2 max but superior reliance on fat oxidation, and maintained O 2 delivery to working muscle. This allows high durability cyclists to maintain a higher relative intensity when fatigued. Durability may be less dependent on peak physiological capacity and more related to metabolic substrate utilization and pacing strategy, since peripheral fatigue was similar between high and low durability groups despite markedly different performance outcomes when fatigued. This study was supported by the Turing Scheme (UK), Mitacs Globalink and NSERC (Canada). This abstract was presented at the American Physiology Summit 2026 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.