The objectives were to (1) prospectively evaluate fatigue utilizing validated instruments in patients with localized prostate cancer, and (2) examine the relationships between fatigue, depression, quality of life, and sleep disturbance. The instruments used included: Piper Fatigue Scale, Beck Depression Inventory, Epworth Sleepiness Scale, and Functional Assessment of Cancer Therapy for Prostate Scale. Data on cancer stage, prostate specific antigen levels, hematocrit, patient's body weight and radiation dosage were recorded. Patients were evaluated preradiotherapy, middle of radiotherapy, completion of radiotherapy, and at 4-5 weeks follow-up. Thirty-six veterans with localized prostate cancer were studied. Mean age was 66.9 years (range 55-79). Duration of treatment was 7-8 weeks. Univariate procedure and Wilcoxon Signed Rank-test were used to examine changes in pretreatment scores for each of the three subsequent study periods. To adjust for multiple comparisons Bonferroni test was used. Spearman Correlations were calculated among parameters. No significant changes were noted in mean scores of hematocrit and body weight during the study period. On the Piper Fatigue Scale, adjusted for multiple comparisons, the median scores were significantly higher at completion of radiotherapy as compared with preradiotherapy values. Three patients (8%) were experienced fatigue according to Piper Fatigue Scale before treatment as compared to nine patients (25%) at completion of radiotherapy. On Prostate Cancer Specific and Physical Well Being subscales of the Functional Assessment for Prostate Cancer Therapy, the scores were significantly lower at middle and completion of radiotherapy than at pretreatment. At preradiotherapy, middle of radiotherapy, completion of radiotherapy and follow-up evaluation, patients scoring higher on the Piper Fatigue Scale were more likely to report a poorer quality of Physical Well Being on Functional Assessment of Cancer Therapy for Prostates. No significant changes were noted in the Beck Depression Inventory and Epworth Sleepiness Scale scores during treatment. Eight patients scored 10 or more on the Beck Depression Inventory before starting radiotherapy, suggesting depressive symptomatology. Of these, only seven patients scored 10 or more at completion of treatment. The incidence of fatigue is lower in our study than in previously published data. A relationship exists between fatigue scores and physical well being subscale scores. Higher scores on the Piper Fatigue Scale at the completion of radiotherapy, as well as no changes on depression and sleepiness scales, suggest that fatigue may not be the result of depression or sleep disturbance. Based upon our previous work, we propose that the physical expression of fatigue may be secondary to a decline in neuromuscular efficiency and enhanced muscle fatigue.