Impact of physical exercise on the kynurenine pathway in patients with cancer: current limitations and future perspectives

With great interest, we have read the current publication of Herrstedt et al. [1], investigating the influence of a chemotherapy-accompanying supervised exercise program on the kynurenine pathway and its association with depressive symptoms in patients with gastro-esophageal cancer.

Overall, the idea of exploring this metabolic branch of tryptophan metabolism in the context of exercise and cancer is a brilliant idea, mainly due to two reasons. First, overactivation of the kynurenine path is closely related to the development and severity of neurological disorders, such as depressions [2] which can frequently be observed in patients with cancer. Second, the mentioned overactivation is linked to the development and progress of several cancer types, through its immunosuppressive properties [3].

Regarding the neurological aspect, the title of the paper suggests a direct association between exercise-induced alterations of the kynurenine pathway and improvements in depressions. In fact, this association has not been shown in the current work. Moreover, analyzing such an association in a population that does not suffer from depressive symptoms is questionable. Of note, ‘depressive symptoms’ which were assessed by the HADS in the current work are not equal to ‘depressions’ (as suggested to be measured in the title). In this context, it could be added that (i) following studies investigating such relations may only include patients reaching a cutoff level of depressive symptoms at baseline and (ii) that further research may include more detailed information on depressions and related symptoms, e.g., fatigue or cancer-related cognitive impairments.

Herrstedt et al. [1] used a broad and intelligent set-up of biological markers (tryptophan metabolites, markers of inflammation, etc.). Nevertheless, the only molecular link between the kynurenine pathway, exercise and depressive symptoms which can be derived from the data is that muscle kynurenine 3-monooxigenase (KMO) is increased in controls after the intervention. However, this result is poorly supported by the data, since it is based on an unadjusted post-intervention comparison of a non-randomized controlled trial with a small sample size (Figure 3). In fact, the few longitudinal data on KMO expression which were shown did not support this finding (Figure 3). GAPDH was used as housekeeping gene for qRT-PCR. The authors mentioned that this gene showed no differential gene expression between the groups (it remains unclear when: at baseline, post-intervention and change). GAPDH has been described to show muscle fiber type-specific expression [4] and has been considered to be inappropriate for normalization in muscle research [5]. Since studies suggest exercise-induced alterations in muscle fiber type composition [6,7], it cannot be ruled out that the expression of this glycolytic enzyme is affected by the exercise intervention itself. The authors stated that a fiber-type shift would be unlikely since a previous resistance exercise intervention did not induce those in patients with germline cancer [8]. However, resistance exercise may consequence in differential adaptions compared to a multimodal exercise program including endurance components. Apart from that, quinolinic acid seems to increase only in controls. Interestingly, these data have not been tested between groups, while raw data suggest an increase in the exercise group as well. A comparison of deltas would have been much more appropriate.

Finally, one important information on the kynurenine pathway is lacking. Further information on the activation of kynurenine pathway could have been provided without assessing more markers. By calculating and presenting the kynurenine/tryptophan ratio, the authors could have shown whether the activation of the pathway is influenced either by chemotherapy, by exercise or by both. Besides the constitutive expression of the liver-specific tryptophan dioxygenase (TDO), inflammation is known to induce its isoenzyme, the indolemanine 2,3-dioxygenase 1 (IDO1) in several tissues [3]. As mentioned above this is highly relevant in the context of cancer not only because of activation of the ‘neurotoxic’ branch of the kynurenine path, but also regarding its ‘immunomodulating’ branch. Currently, several IDO Inhibitors are tested as immunotherapies in clinical trials [9].

In conclusion, we highly appreciate the work of Herrstedt and colleagues. However, in our opinion, the results are partially overinterpreted considering all methodological limitations mentioned above.

Disclosure statement

No potential conflict of interest was reported by the authors.