Page 134 - An Evidence Review of Active Surveillance in Men With Localized Prostate Cancer

Comparison Between Observational Management Strategies and Radiation Therapy
Detailed results from the eight studies that compared observational management strategies
and radiation therapy (RT) are shown in Appendix Table C4.3. No study reported on the effects
of treatment on the development of metastatic disease.

225
Randomized controlled trial. One study reported quality of life comparisons based on a
subset of 54 surviving responders from the UMEA1 trial. UMEA1 was a randomized trial of 166
patients comparing RT with WW. The authors stated that the complete trial data are “not yet
ready for publication.” At 10 years of followup, the study reported no statistically significant
differences in health-related quality of life function or symptom scales. No significant difference
was observed between RT (mean=2.3) and WW (mean = 2.3; P not reported) for “limitations in
daily life” caused by prostate cancer. No significant difference was observed for “life situation”
between RT (mean=3.7) and WW (mean=3.3; P=0.398). Out of 20 comparisons in
symptom/function scores (abstracted in Appendix Table C4.3) only that of “sexual bother” was
statistically significant (mean scores: 3.8 vs. 7.4 in the WW and RT groups, respectively; higher
values indicate worse function; P=0.011). A weak urinary stream was the most reported
symptom and the mean score differed significantly between WW (mean 4.8) and RT (mean 3.0;
higher values indicate worse function; P = 0.034). No differences were observed at 10 years
between RT and WW regarding the questions “How would you feel if you lived the rest of your
life with your urinary/bowel/sexual problems as they are now?” (P values comparing the two
groups were 0.643, 0.653 and 0.819, respectively).

Observational studies—Cancer-specific mortality. One study using data from NCRSFS
reported that RT was associated with a non-significant improvement in prostate cancer-specific
mortality compared to surveillance (WW or AS in aggregate; HR 0.70; 95 percent CI 0.45, 1.09)
at 8.2 years of followup.

Observational studies—All-cause mortality. Two studies reported adjusted estimates
comparing all-cause mortality between observational management strategies and RT. 208,223 One
study was based on the SEER-Medicare database, 208 and the other used NCRSFS data. 223 Both
studies reported significantly lower mortality rates in patients treated with RT compared with
patients managed with an observational strategy (HR 0.81; 95 percent CI 0.78-0.85, 208 and HR
0.68; 95 percent CI 0.57, 0.82, 223 respectively).

Observational studies—Morbidity of primary treatment. One report analyzed the incidence
of treatment for urethral stricture captured in the CaPSURE registry and did not find a significant
difference between patients on WW and patients treated with EBRT, or between patients on WW
and patients treated with BT over a median followup of 2.7 years. 218

Observational studies—Quality of life. Three studies reported quality of life outcomes. 224,226,229
From the CaPSURE registry, patients on WW had significantly lower mean SF-36 score in the
social function domain than patients treated with RT (89 vs. 86; P < 0.05, respectively), but no
229
difference was found in three other domains of SF-36. In a study of 4 academic medical
centers from Wisconsin no significant difference in disease-specific and general QoL was
observed between patients managed with RT and those managed with expectant management. 226
Finally, a study based on the Eindhoven Cancer Registry found that RT had a negative effect on

80