Resistance training load does not explain interindividual variability in muscle hypertrophy: insights from recent evidence

Muscle strength and hypertrophy are highly valued adaptations to resistance training (RT), and several training variables have been examined to optimize these outcomes. Despite heavy loads benefiting strength gains (Schoenfeld et al., 2017), considerable debate remains regarding the relative contributions of the magnitude of external load and intrinsic biological factors in mediating hypertrophic responses. Experimental and meta-analytic evidence has demonstrated that both high-load and low-load RT can promote comparable hypertrophy when training to momentary muscular failure (Schoenfeld et al., 2017). In light of this ongoing discussion, a recent study published in The Journal of Physiology by Lees et al. (2026) provides new and relevant evidences that contributes meaningfully to this field of inquiry. This study investigated whether the magnitude of RT load and limb location (upper vs. lower body) mediate muscle hypertrophy when effort is matched to volitional fatigue. Using a unilateral, within-subject design, untrained healthy young men completed 10 weeks of RT in which each limb was randomized to either a higher-load (HL; 8–12 repetitions at ∼70–80% 1 repetition maximum (1RM)) or lower-load (LL; 20–25 repetitions at ∼30–40% 1RM). Muscle size was assessed through dual-energy X-ray absorptiometry (DXA)-derived appendicular lean mass, ultrasound-derived muscle thickness and cross-sectional area, and muscle fibre cross-sectional area from vastus lateralis biopsies. Interestingly, muscular adaptations of each participant were ranked using a composite measure considering these distinct outcomes. Traditionally, labelling participants in RT interventions according to their responsiveness levels (e.g. responders vs. non-responders; high vs. low-responders) considers one specific measure of skeletal muscle hypertrophy. However, labelling responsiveness considering only one specific measure of skeletal muscle hypertrophy may provide a limited or incomplete representation of an individual's adaptive profile, since a misalignment between macroscopic (e.g. muscle thickness; muscle cross sectional area) and microscopic (e.g. muscle fibre cross-sectional area) assessments of hypertrophic outcomes has been previously reported (Haun et al., 2019). Therefore, by adopting this multi-assessment approach, Lees et al. (2026) contribute with relevant knowledge toward a better understanding of the responses to different loading schemes, as well as the heterogeneity of muscular adaptations to distinct RT programmes. It is also important to note that most intervention studies investigating individual-level muscle hypertrophic responsiveness typically assess only a single site or muscle group. Although this approach is valid, it may raise questions about whether the magnitude of individual responsiveness is, to some extent, muscle group-dependent. To the best of our knowledge, Lees et al. (2026) are the first to consider distinct muscle groups (i.e. vastus lateralis and biceps brachii) when assessing the heterogeneous adaptations at both the between- and within-subject levels. Consistent to the initial hypothesis, the authors reported that: (i) hypertrophic adaptations were highly heterogeneous between individuals, aligning with classic work reporting substantial variability across individuals even when RT protocols are tightly controlled (Hubal et al., 2005); and (ii) muscle size responses were relatively conserved within individuals across upper- and lower-limb muscles and distinct loading conditions. This intraindividual conservation, evidenced by greater shared variance within than between participants, supports previous findings suggesting that endogenous biological factors, rather than external load manipulation, are the primary determinants of hypertrophic responsiveness (i.e. subjects considered as high responders to a given training programme tend to present a similar response independently of the muscle group and loading pattern adopted). In order to better understand the phenotypic responses induced by the distinct loading schemes, Lees et al. (2026) adopted an integrated approach by incorporating mechanistic-based data. Using deuterated-water labelling, myofibrillar protein synthesis (MyoPS) rates were reported to be elevated during the early phase of RT (i.e. first week) relative to rest and were attenuated by the 10th week. This attenuated response of MyoPS observed is consistent with prior research showing that early elevations in protein synthesis may primarily reflect repair and remodelling associated with exercise-induced muscle damage and, as the repeated-bout effect occurs, the MyoPS response shifts to become more closely aligned with true myofibrillar accretion (i.e. muscle hypertrophy) (Damas et al., 2016). Additionally, Lees et al. (2026) also described that MyoPS did not differ between HL and LL conditions at either time point and, similar to hypertrophy outcomes, showed less variability within individuals than between individuals. Average volume load over the 10 weeks did not differ between HL and LL for the legs (LL: 81,968 ± 26,924 kg; HL: 78,583 ± 22,984 kg; P = 0.197). In contrast, arms showed a higher average volume load in the HL condition (58,756 ± 12,351 kg) than in the LL condition (38,334 ± 8496 kg; P 0.05), which does not corroborate previous literature indicating that heavier loads confer greater strength gains (Schoenfeld et al., 2017). Finally, there was minimal shared variance between changes in muscle size and strength. Collectively, this study adds relevant insights to the current body of evidence when indicating that the magnitude of RT load may not be the major determinant for the muscle strength and hypertrophic responses in untrained healthy young men. Rather, these data suggest a partially conserved hypertrophic phenotype within individuals, as indicated by shared variance across upper and lower limbs and across load conditions, while the magnitude of external load did not meaningfully modulate hypertrophy outcomes. The substantial interindividual heterogeneity consistently observed across studies suggests that adaptive potential is not uniform across trainees and may represent a relatively conserved phenotypic trait within each individual (Rantila et al., 2025). In light of this, and acknowledging that individual responsiveness cannot be easily predicted or assessed a priori, we propose that RT prescription should adopt a more flexible and integrative framework across the training cycles. Rather than prioritizing a single loading paradigm, practitioners may benefit from incorporating multiple distinct mechanical stimuli, such as variations in load, repetition range, contraction velocity, proximity to failure, range of motion and rest intervals, with the aim of exposing the neuromuscular system to diverse adaptive stimuli. This approach may better accommodate biological variability, facilitate the expression of individual hypertrophic potential and ultimately provide a more individualized pathway for optimizing RT-induced muscle adaptations. An important point for discussion is the study population. In the study by Lees et al. (2026), all participants were recreationally active. Therefore, whether similar responses would be observed in distinct populations (e.g. sedentary individuals; advanced lifters; older adults) remains unclear and warrants further investigation, given its significant clinical and practical implications. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. The authors declared no conflict of interest. P.G.S.B: Conception or design of the work, drafting the work or revising it critically for important intellectual content. J.B.B.C and B.N.I: drafting the work or revising it critically for important intellectual content. All authors have read and approved the final version of this manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed. P.G.S.B. receives a postdoctoral fellowship (CNPq, grant number 88887.102383/2025-00). J.B.B.C. receives a doctoral fellowship (FAPESP, grant number 2024/14207-0).