Beyond erythropoietin: What hormonal profiling reveals about the true cost of high‐altitude training

Altitude training camps have become standard practice in elite endurance sports, primarily targeting increases in total haemoglobin mass (Hbmass) to enhance oxygen transport capacity. The rationale is straightforward, following a simple input–output system: hypoxic exposure stimulates erythropoietin production, which in turn increases Hbmass and oxygen delivery capacity to working muscles. However despite decades of research the haematological response to altitude training remains remarkably heterogeneous. Although some athletes respond robustly, others show minimal gains in Hbmass. Moreover although a dose–response relationship exists at the group level – that is, higher hypoxic dose generally correlates with greater Hbmass increases (Millet et al., 2019) – this pattern does not hold at the individual level. To gain further insight into the mechanisms responsible for this variation Faiss and colleagues (2025) addressed in a recent article published in The Journal of Physiology, whether endocrine and steroidal markers might explain the heterogeneous response to altitude training (Faiss et al., 2025). Faiss and colleagues monitored 31 professional male cyclists during 3 week altitude training camps at either 2050 m (n = 19, live-high train-low) or 3000 m (n = 12, live-high train-high). Blood samples were collected before, during and after altitude exposure to measure haematological parameters alongside steroidal (testosterone and androstenedione) and endocrine hormones (e.g. insulin-like growth factor-1 (IGF-1)). Despite the 3000 m camp providing a 67% higher hypoxic dose (1500 km·h−1 vs. 900 km·h−1) Hbmass increased similarly in both groups (∼3.5%). This finding challenges conventional dose–response assumptions and suggests that higher altitude may not necessarily yield greater haematological adaptation. From a practical standpoint the hormonal findings offer valuable insights for coaches, scientists and athletes. Testosterone concentration increased substantially at 2050 m, while remaining stable at 3000 m, despite similar Hbmass gains. Furthermore testosterone concentrations showed a moderate positive correlation with changes in Hbmass across the camps. Whether this suppression at 3000 m resulted from the altitude itself, inadequate energy availability (evidenced by body weight loss) or the compounded stress of living and training at high altitude remains to be determined. Regardless this observation underscores the importance of maintaining an anabolic hormonal state during altitude training. This requires careful monitoring of nutrition, training load and physiological recovery markers (e.g. heart rate variability, resting heart rate) throughout altitude camps. Collectively the findings suggest that moderate altitude (∼2000 m) with a live-high train-low approach may represent an optimal strategy, providing sufficient erythropoietic stimulus while preserving anabolic hormonal status and allowing higher training intensities at lower altitude. Athletes showing signs of maladaptation may therefore benefit from reducing altitude exposure or training load, rather than simply persisting with the prescribed altitude camp structure. Beyond testosterone, the IGF-1 findings add another layer of complexity to altitude adaptation. IGF-1 concentrations remained stable during altitude exposure but increased slightly upon return to sea level. This delayed response aligns with the earlier observation that combined hypoxia and exercise disrupt the growth hormone (GH)–IGF-1 axis (Schmidt et al., 1995). The high energy demand during hypoxic exercise may override the GH-mediated stimulus on hepatic IGF-1 production. The postaltitude elevation of IGF-1 likely represents hormonal recovery once the hypoxic stress is removed and may play a regulatory role in the normalisation of erythropoiesis upon return to sea level. This interpretation is supported by the negative correlation between IGF-1 and Hbmass (r = −0.25). Whether the magnitude or timing of the IGF-1 response affects how long Hbmass elevations persist, which is critical for optimising competition scheduling postaltitude, remains an important question for future research. Nonetheless although significant correlations were reported between hormonal markers and changes in Hbmass, the modest variance explained (testosterone R2 = 0.14, IGF-1 R2 = 0.06) suggests that these markers have limited utility for individualising altitude training prescription in athletes. Other hormonal pathways not assessed in this study may play more important roles. For instance, appetite-regulating hormones warrant investigation given the body weight loss observed at 3000 m, as inadequate energy availability is likely to contribute to impaired adaptation. Similarly, biomarkers previously associated with non-functional overreaching, such as growth differentiation factor 15 (GDF-15), could help identify when hypoxic stress becomes counterproductive (Poffé et al., 2023). Future research employing multiomics approaches (transcriptomics, proteomics and metabolomics) may provide a more comprehensive understanding of which physiological processes change during altitude training and how these responses differ across altitude levels. Such broad molecular profiling may ultimately identify biomarker panels capable of predicting individual responses to altitude training more effectively than single hormonal measurements. An additional consideration is the marked difference in plasma volume contraction between altitude levels. Plasma volume decreased by 6% at 2050 m and by 12% at 3000 m. Although no correlation emerged between plasma volume changes and Hbmass gains, this haemoconcentration may nonetheless play an important role in altitude adaptation. Previous studies showing Hbmass increases following active and passive heat training, which expands plasma volume, suggest that changes in blood volumes impact erythropoietic responses (Lundby & Robach, 2025). Theoretically the greater plasma volume contraction at 3000 m could have reduced the hypoxic stimulus by preserving arterial oxygen content despite lower ambient oxygen pressure, thereby attenuating EPO production and Hbmass gains. This may partially explain why the higher hypoxic dose at 3000 m failed to produce greater erythropoietic adaptation. Therefore it would be interesting to investigate whether interventions that minimise plasma volume contraction, such as (hyper)hydration protocols and heat acclimation, could improve Hbmass gains to altitude training. Furthermore, given the plasma volume contraction inherent to altitude training, heat training may represent a more practical alternative for many athletes. Beyond increasing Hbmass, heat acclimation provides directly transferable benefits for competition in warm conditions. Moreover heat training can be easily conducted at home without the need for expensive altitude camps, while other vital parameters, such as sleep quality and training intensity, are likely to be less compromised than with altitude exposure. A notable limitation is that although the study was motivated by individual variability in altitude responses, it primarily reports group-level statistics. The standard deviations suggest substantial interindividual variation, yet although individual data points are shown, they are not connected across timepoints, preventing assessment of individual trajectories. Furthermore responder/non-responder analyses could have assessed whether baseline parameters or early hormonal responses predict individual adaptation to altitude. Notwithstanding these limitations the inclusion of professional athletes represents a major strength, as this is the population that will benefit most from a better understanding of individual responses to altitude training camps. The authors appropriately acknowledge that the observational design limits causal inference. Although the inclusion of professional athletes enhances ecological validity and relevance to professional athletes, it restricts the use of randomised altitude assignments and matched control groups, which are essential for establishing causality. Future controlled studies with randomised altitude assignments and matched control groups should therefore be conducted in well-trained recreational athletes to establish causal relationships between hormonal responses and haematological adaptation. Such findings could then inform more targeted monitoring strategies for professional athletes during altitude training camps. Faiss and colleagues have provided valuable insights into the hormonal landscape of altitude training, revealing that pursuing higher Hbmass gains comes with anabolic costs. The study suggests that moderate altitude may offer an optimal balance between a sufficient hypoxic stimulus for erythropoiesis and anabolic hormonal status. However, the modest correlations between the measured hormones and Hbmass indicate that simple hormonal monitoring cannot yet guide personalised altitude training. Moving forward, controlled studies integrating hormonal profiling, physiological markers of recovery/overreaching and omics analyses may be necessary to establish causal mechanisms and better understand the individual variation to altitude. Until more predictive tools emerge the practical advice for coaches and athletes is to carefully monitor training load, hydration status, recovery markers and energy balance while adopting a live-high, train-low approach at moderate altitude (2000–2500 m). 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. No conflicts of interest, financial or otherwise, are declared by the authors. All authors have read and approved the final version of this manuscript and agreed 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. Siemon Vermeiren is supported by Hasselt University Special Research Fund (BOF) grant BOF25OWB33. Nathan Vermaerke is supported by BOF grant BOF25DOCLI08.