Key Question 3

We identified two additional trials (n=43 and 63) of adjuvant intravesical therapy versus TURBT alone.^{1, 2} One trial was rated medium risk of bias¹ and the other high risk of bias.² Both trials evaluated single dose therapy with MMC in patients with lower-risk NMIBC.

We updated the meta-analyses of MMC versus TURBT alone with the new trials. For recurrence, we also added data from a previously included trial that was omitted in the original report.³ The only outcomes that were affected in the new analysis were bladder cancer recurrence and progression; updated estimates and conclusions were generally similar to those in the original report (Table 1).

Table 1

Updated results for MMC versus no intravesical therapy (TURBT alone).

Regarding head-to-head comparisons of intravesical therapies, we identified one additional trial (n=407, medium risk of bias) of intravesical BCG versus BCG plus MMC given sequentially.⁴ We updated the meta-analyses with the new study; we also added data on progression from a previously included trial⁵ that was omitted from the original report. Updated estimates and conclusions were similar to those in the original report (Table 2)

Table 2

Updated results for BCG versus BCG plus MMC given sequentially.

For comparisons involving dose or duration, we identified one additional trial (n=166, medium risk of bias) that compared BCG Tokyo strain 40 mg versus 80 mg (each administered once weekly for eight weeks).⁶ It was consistent with two trials in the original report that found no clear difference between higher and lower doses of BCG. In the new trial, there was no clear difference between the lower and higher dose in risk of complete response (78% vs. 85%, p=0.12), recurrence-free survival (p=0.94), or progression (6.2% vs. 5.8%).⁶

Key Question 6

We identified one new trial (n=85, medium risk of bias) of fluorescent cystoscopy plus white light cystoscopy versus white light cystoscopy alone that reported effects on bladder cancer recurrence.⁷ It found no clear difference between fluorescent plus white light cystoscopy versus white light cystoscopy alone in risk of recurrence through up to 40 months of follow-up (38% vs. 46%, RR 0.82, 95% CI 0.49 to 1.35), though results slightly favored fluorescent cystoscopy. All patients received a single instillation of epirubicin immediately following TURBT. There were also no differences between fluorescent versus white light cystoscopy when patients were stratified according to tumor risk category. The trial did not evaluate effects on progression or mortality. Patients were randomized to intravesical HAL or placebo administered one hour prior to white light cystoscopy, which was performed in all patients. Surgeons were blinded to treatment allocation until after white light cystoscopy was completed; only patients randomized to HAL underwent fluorescent cystoscopy.

We updated the meta-analyses on recurrence with data from the new trial. We also found errors in the data used for long-term recurrence and progression for one trial⁸ included in the original report and used corrected data in the updated meta-analysis. The updated analyses resulted in similar estimates and conclusions and are shown in Table 3. Estimates and overall conclusions were similar to those in the original report.

Table 3

Updated results for fluorescent cystoscopy versus white light cystoscopy.

Study characteristics and risk of bias ratings for the additional studies are shown in Tables 4, 5 and 6.

Table 4

Characteristics and results of additional trials of intravesical therapy.

Table 5

Characteristics and selected results of an additional trial of fluorescent cystoscopy.

Table 6

Risk of bias of additional trials.