Associative learning contributes to the persistence of fatigue-like behavior in male mice in a model of cancer survivorship

Abstract

Persistent fatigue is a debilitating side effect that impacts a significant proportion of cancer survivors for which there is not yet an FDA-approved treatment. While certainly a multi-factorial problem, persistent fatigue could be due, in part, to associations learned during treatment. Therefore, we sought to investigate the role of associative learning in the persistence of fatigue using a preclinical model of cancer survivorship. For this purpose, we used a murine model of human papilloma virus-related head and neck cancer paired with a curative regimen of cisplatin-based chemoradiation in male C57BL/6J mice. Fatigue-like behavior was assessed by measuring variations in voluntary wheel running using a longitudinal design. Treatment robustly decreased voluntary wheel running, and this effect persisted for more than a month posttreatment. However, when wheels were removed during treatment, to minimize treatment-related fatigue, mice showed a more rapid return to baseline running levels. We confirmed that the delayed recovery observed in mice with continual wheel access was not due to increased treatment-related toxicity, in fact running attenuated cisplatin-induced kidney toxicity. Finally, we demonstrated that re-exposure to a treatment-related olfactory cue acutely re-instated fatigue. These data provide the first demonstration that associative processes can modulate the persistence of cancer-related fatigue-like behavior.

Keywords: Associative learning; Cancer; Chemoradiation; Conditioned responses; Fatigue; Survivorship.

Figure 1.. Mice show a delayed recovery of voluntary activity following a curative regimen of cancer therapy.

Mice were injected with tumors (n = 14) or vehicle (n = 6) at day 0 and then started a curative regimen of cisplatin-based chemoradiation (CRT) or sham treatment (A). Mice undergoing treatment showed a reduction in body weight (B) and a complete tumor response (C). Treated mice also displayed a deficit in voluntary wheel running data that persisted for more than one month after treatment completion (D). Using the dynamic weight bearing (DWB) task it was demonstrated that treatment did not impact weight bearing on the restrained (E). At the time of study end, liver Il6 mRNA expression was still elevated, but all other observed inflammatory and metabolic markers were different from healthy control mice (F). * p< 0.05, *** p < 0.001.

Figure 2.. Lack of wheel access during treatment accelerates recovery of post-treatment voluntary activity.

Mice were injected with tumors on day 0 and then started a curative regimen of cisplatin-based chemoradiation (CRT) or sham treatment either with or without wheels removed (n = 5 −6 mice/group) (A). Mice undergoing treatment showed an equivalent treatment-related weight loss in both wheel and no wheel conditions (B). All but one mouse showed a complete tumor response, the non-responder was excluded from analyses (C). During the treatment phase, tumor-bearing mice displayed fatigue-like behavior. In survivorship, when all mice had wheels returned, mice who had their wheels removed during treatment (no wheel condition) had improved running compared those with continuous wheel access (D). *** p < 0.001.

Figure 3.. Wheel access did not exacerbate cisplatin-induced kidney toxicity.

Mice were injected with tumors or vehicle at day 0 and then started a curative regimen of cisplatin-based chemoradiation (CRT) or sham treatment. Tissue was collected on day 25 (A). Mice undergoing treatment showed an equivalent treatment-related weight loss in both wheel and no wheel conditions (B). All mice showed a complete tumor response (C). During treatment mice showed the expected reduction in voluntary wheel running (D). For those without wheels we monitored bins of nightly home cage activity, which also showed a significant decline in response to treatment (E). The wheel running mice did not show an increased kidney toxicity measured by qPCR markers in the kidney (F), plasma urea (G), and plasma creatinine (H). In fact, the running group showed blunted toxicity profiles across several of these measures. n = 7 –8 mice/group. * p< 0.05, *** p < 0.001.

Figure 4.. Re-exposure to a treatment-associated olfactory cue acutely reinstated deficits in voluntary activity.

Mice were injected with tumors at day 0 and then started a curative regimen of cisplatin-based chemoradiation (CRT) (A). Mice undergoing treatment showed the expected reductions in body weight (B), tumor response (C), and voluntary activity (D) during treatment. The previously paired olfactory cue or an odorless control was added back to the cage 2 –3 weeks after treatment completion in a counterbalanced fashion. Presentation of the cue associated with treatment significantly decreased running the night immediately following re-exposure with recovery by the second night (E). N = 17. *** p < 0.001