Reply from J. E. Mallinson, D. Constantin‐Teodosiu, D. Constantin, M. V. Franchi and P. L. Greenhaff: NSAIDs and muscle hypertrophy – preconceptions and differing interpretations in the face of a paucity of data from long‐term healthy volunteer studi

We would like to thank the authors of the recent Letter (Mandić et al. 2026) for taking the time to read and review our work (Mallinson et al., 2026). The Introduction of Mallinson et al. (2026) addressed in detail the widespread misuse of non-steroidal anti-inflammatory drugs (NSAIDs) in sport, which was a catalyst in the aetiology of the project, particularly given that longitudinal, interventional studies addressing the impact of NSAID administration on resistance exercise-induced muscle adaptations in healthy volunteers are limited in number. As might be expected, therefore, we reported the findings of the authors of the Letter (Mandić et al. 2026) that 8 weeks of ibuprofen ingestion (1200 mg/day) attenuated strength and hypertrophic responses to resistance training in young, healthy adults compared to acetylsalicylic acid ingestion (75 mg/day; Lilja et al., 2018). These observation of Lilja et al. (2018) were conveyed in Mallinson et al. (2026) (diclofenac 75 mg/day for 12 weeks) alongside those of Trappe et al. (2011), highlighting that oral administration of either paracetamol (4000 mg/day) or ibuprofen (1200 mg/day) to healthy, older volunteers during 12 weeks of resistance training augmented muscle volume gains compared to placebo intervention. Given that the findings of Mallinson et al. (2026) and Trappe et al. (2011) concluded that NSAID ingestion augmented training-induced muscle hypertrophy, we find the claim of Mandić et al. (2026) that the findings were ‘apparently contradictory’ to the literature to be overstated. Similarly, the claim that Mallinson et al. (2026) ‘declined a reviewer request for further comparison with Lilja et al. (2018)’ is incorrect. As the record shows, it is stated that the authors would be ‘guided by the Reviewer's opinion’. Given Lilja et al. (2018) compared the impact of 8 weeks high dose ibuprofen vs. low dose acetylsalicylic acid ingestion, with no placebo intervention, we preferred not to be critical of a study felt not to be directly comparable to the placebo controlled studies of Trappe et al. (2011) and Mallinson et al. (2026). The Reviewer agreed with this interpretation. Moreover, the discussion of Lilja et al. (2018) acknowledges this study limitation. This is not to say that the observations of Lilja et al. (2018) are not valuable contributions to the literature, particularly given the paucity of data available involving chronic longitudinal study designs. Turning to other points raised in the Letter, Mandić et al. (2026) are of the opinion that the placebo intervention in Mallinson et al. (2026) produced ‘only a modest response’ from the perspective of muscle volume and strength gains. However, Mallinson et al. (2026) reports quadriceps muscle cross sectional area (CSA) was increased by 4.3% above baseline after 28 days of resistance training (P = 0.039), which we would not describe as modest. For example, Seynnes et al. (2007) reported a 5.1% increase in MRI-derived mid-quadriceps CSA following 20 days of resistance exercise training also performed three times per week. Furthermore, there was a trend for muscle CSA to remain increased above baseline in the placebo group at 84 days (3.9%, P = 0.054), which was no different to the increase recorded at 28 days (P = 0.999). What was reported therefore in Mallinson et al. (2026) was a plateau in muscle CSA gains in the placebo group beyond 28 days of training. The manuscript discussion acknowledges this was unexpected and could have been a consequence of low statistical power, particularly given volunteers in the placebo group performed the same volume and load of training as volunteers in the NSAID group, who did not experience a plateau in muscle CSA gains from 28 to 84 days (Mallinson et al., 2026). The discussion also points out that increases in muscle mass (Ogasawara et al., 2013) and post-exercise muscle protein synthesis (Brook et al., 2015) have been reported to plateau with chronic resistance training, and goes on to highlight that differences in literature reports are likely attributable to variances in the duration of resistance exercise training, the type of contractions performed, the intensity and the total number of contractions executed per bout of training and the type and dose of NSAID administered. The same points apply to the muscle strength responses in the placebo group. The failure of resistance exercise training to produce further strength gains in trained volunteers is not uncommon, often referred to as a training plateau, which may have been a factor in the study of Mallinson et al. (2026) that recruited trained male volunteers into the study to minimise neural responses to resistance exercise (Moritani, 1993), and differs from the study of Lilja et al. (2018) that recruited recreationally active men and women. We do not accept the comment of Mandić et al. (2026) that ‘the choice of training protocol may partly explain this attenuated response’, with the implication that the concentric training protocol described in Jones et al. (2004) and Mallinson et al. (2026) may have been an insufficient stimulus to produce training adaptation in the latter study. Contrary to this perspective, the same (or very similar) isokinetic concentric-only protocols have been shown to result in comprehensive quadriceps muscle fibre recruitment (Tesch et al., 1990), muscle transcriptional responses (Murton et al., 2014) and increased muscle protein synthesis (Shur et al., 2024). Furthermore, as outlined above, the volunteers in the placebo group of Mallinson et al. (2026) performed the same volume and load of training as volunteers in the NSAID group, who had a greater hypertrophic response. This leads us to conclude that the training stimulus was sufficient. Indeed, if anything, the total of 5400 maximal, isokinetic, concentric, knee extension contractions over 12 weeks in the placebo group may have been disruptive to adaptation, which the NSAID intervention overcame. This is supported by the Ingenuity Pathway Analysis of muscle mRNA expression that predicted the inhibition of muscle formation in the placebo group at 84 days (Fig. 5A), which contrasted with the high confidence prediction of activation of muscle formation in the NSAID group at 28 days (Fig. 5B) and 84 days (Fig. 5C) in Mallinson et al. (2026). We can confirm the NSAID group performed the same bi-lateral training as the placebo group. The study outcomes for the concentrically trained limbs are reported in Mallinson et al. (2026), and data from combined eccentric–concentric exercise protocols will be reported separately in due course in the context of high-load eccentric exercise induced muscle disruption. This will add further insight regarding the interaction between NSAIDs and muscle inflammation during chronic eccentric muscle loading, which is separate from the stated aims of Mallinson et al. (2026) where muscle pro-inflammatory protein expression levels were no different between groups at any time point. We concur that the statistical outcomes for hypertrophy and strength for the placebo group differ between Mallinson et al. (2020) and Mallinson et al. (2026), which is due to the different statistical approaches employed. Mallinson et al. (2020) compared concentric vs. combined concentric/eccentric training in the same volunteers utilising a within subject design. Mallinson et al. (2026) required between subject comparisons, which is associated with inherently greater variability. Mallinson et al. (2026) did not ‘suggest that diclofenac's preferential COX-2 inhibition explains their findings’. Rather the discussion of Mallinson et al. (2026) asserts that variation in NSAID selectivity and the dose administered may explain differences in outcomes across studies. This positioned the findings of Mallinson et al. (2026) within the broader published literature, and in doing so highlighted several studies. Irrespective of this point, however, longitudinal, interventional study designs that have addressed the impact of NSAID administration on resistance exercise-induced muscle adaptations in healthy volunteers (irrespective of the volunteer age or COX inhibitor selectivity) are very limited in number, and the mechanistic basis of reported outcomes are poorly understood. Indeed, the available research is both limited and conflicting. For example, NSAID ingestion has been reported to attenuate (Lilja et al., 2018), have no effect (Krentz et al., 2008) or augment (Mallinson et al., 2026) hypertrophic responses to 6–12 weeks of resistance exercise training in healthy, young volunteers. Seemingly at variance with the stance of Mandić et al. (2026), it is currently unclear why there are disparate effects on the muscular response to resistance exercise training with chronic NSAID use. The study of Mallinson et al. (2026) is therefore a useful addition to the field. Mandić et al. (2026) claim that the disconnect between muscle hypertrophy and strength gains is notable in Mallinson et al. (2026). However, as highlighted in the Discussion of Mallinson et al. (2026), such disconnect is not uncommon in studies examining the impact of pharmacological interventions on muscle mass and functional gains. For example, bimagrumab, a human monoclonal anti-ActRII antibody, has been shown to be highly effective at promoting muscle hypertrophy in humans, but does not augment muscle functional gains, even in healthy, young volunteers (Rooks et al., 2017). Indeed, Lilja et al. (2018) report that muscle volume increased more with acetylsalicylic acid (8.1%) than with ibuprofen (3.7%) in the weight-stack trained limb (P = 0.035), but no group differences were noted for strength gains (P = 0.537), which supports the premise of a disconnect, even in their own data. As Mandić et al. (2026) convey, and Mallinson et al. (2026) acknowledge, NSAID-induced water retention may have contributed to the MRI-derived muscle volume increases reported in Mallinson et al. (2026). However as outlined above, based on longitudinal changes in muscle mRNA expression in each study group, Ingenuity Pathway Analysis predicted the activation of muscle formation in the NSAID group at 28 and 84 days in Mallinson et al. (2026). This contrasted with the predicted inhibition of muscle formation in the placebo group at 84 days, suggesting that the greater hypertrophy in the NSAID group was unlikely to be attributable to water retention. In closing, we thank the authors for their Letter and for giving us the opportunity to further discuss our findings. On reflection, we feel many of the points raised in the letter had already been addressed in Mallinson et al. (2026), but we also recognise that science is not ‘absolute’ in its objectivity, but rather ‘achieves’ objectivity through debate, interpretation and refinement. We would hope the authors of the Letter to the Editor agree that evidence regarding the impact of long-term NSAID use on muscle strength and hypertrophy gains from resistance training is currently both sparse and inconsistent. 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 have no conflicts of interest. P.L.G., J.M.: Conception or design of the work. P.L.G., J.M., M.F., D.C., T.C.: Acquisition, analysis or interpretation of data for the work. P.L.G., J.M., M.F., D.C., T.C.: Drafting the work or revising it critically for important intellectual content. All authors approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship. This work was supported by a grant from the Versus Arthritis (21595), Medical Research Council (MR/P021220/1), and BBSRC (BB/I020713/1).