WASHINGTON — Adding immunotherapy to standard treatment for non-muscle invasive bladder cancer (NMIBC) achieved complete response (CR) in more than 90% of patients who had refused cystectomy for high-risk disease, a small prospective study showed.
Only three of 37 patients did not have a CR at 6 months with pembrolizumab (Keytruda) and Bacille Calmette-Guérin (BCG). After a median follow-up of 22 months, no patient had progressed to MIBC or developed metastasis. Four patients had high-grade recurrence. At 12 months, 94% of the patients were disease free with an intact bladder.
Most patients had treatment-related adverse events (TRAEs), including severe (≥3) immune-related AEs in a fifth of patients, but the safety profile was consistent with previous studies of an immunotherapy-BCG combination, reported Eugene Pietzak, MD, of Memorial Sloan-Kettering Cancer Center in New York City, at the American Urological Association (AUA) annual meeting.
“We believe our findings support pembrolizumab plus BCG as a potential bladder-preserving option in carefully selected patients who decline cystectomy but are at very high risk for progression with BCG alone,” said Pietzak.
The pembrolizumab-BCG combination joins a growing number of regimens vying for a role in treating high-risk NMIBC. Within the past year, three phase III randomized trials have been reported, each involving a different PD-L1 inhibitor.
The CREST trial of sasanlimab and BCG met the primary endpoint of improving event-free survival (EFS) versus BCG alone. The POTOMAC trial of durvalumab (Imfinzi) and BCG showed significant improvement in disease-free survival, and an update reported at the AUA meeting showed a reduction in early recurrence. The ALBAN trial with atezolizumab (Tecentriq) and BCG failed to show an improvement in EFS versus BCG alone, leading investigators to conclude that “any benefit from checkpoint-BCG therapy may be context- and agent-specific rather than a class effect.”
Collectively, the three randomized trials “suggested that the addition of an immune checkpoint inhibitor (ICI) to BCG may reduce the risk of noninvasive recurrences, but with a substantial increased risk of treatment-associated toxicity,” said Pietzak. When used in a relatively unselected population, an ICI plus BCG offers a suboptimal benefit-to-risk balance for most patients with NMIBC.
Distinct from the phase III trials, the current phase II study focused on patients with “very high risk” NMIBC, T1 plus carcinoma in situ. That category is at the very end of the NMIBC risk spectrum and behaves more like MIBC, including an increased risk of understaging and potential to harbor micrometastasis. The three randomized trials enrolled few patients in the very-high-risk category, and the BCG control arm in the studies had low rates of disease progression, said Pietzak.
Most clinical guidelines recommend upfront cystectomy as the preferred treatment for very high-risk T1 NMIBC, he continued. However, cystectomy represents overtreatment for 50-75% of patients and BCG alone represents undertreatment for 25-50%.
“Thus, there is a major unmet need for a treatment strategy that better balances the risks of overtreatment and undertreatment,” said Pietzak.
Whether adding an ICI to BCG achieves that balance remained unclear after the three randomized trials, he added.
Eligible patients for the phase II trial had no prior BCG exposure, histologically confirmed high-grade T1N0M0 urothelial cancer with concomitant CIS, and at least one additional adverse feature. Immediate cystectomy had been recommended in all cases, but the patients declined.
Treatment consisted of IV pembrolizumab every 6 weeks for a maximum of nine doses. Intravesical BCG induction began 3 weeks after the first pembrolizumab dose, with subsequent maintenance doses of BCG at 3 6, and 12 months.
The primary outcome was CR at 6 months. A CR rate ≤50 would be considered inadequate for further investigation, and a rate ≥70% was considered clinically meaningful.
The patient population had a median age of 67, and men accounted for 86% of the patients. All of the patients had high-grade T1 disease and CIS, and 26 had diffuse or multifocal CIS. Three-fourths of the patients had at least two adverse features and half had three or more, the most common being tumor size >3 cm, extensive lamina propria invasive, multiple T1 tumors, and deep lamina propria invasion.
The trial met the primary outcome, showing a 92% rate of clinical CR at 6 months. The four high-grade recurrences consisted of two cases of recurrent high-grade T1 and one each of bladder CIS/Tis and prostatic urethral CIS. Two patients underwent cystectomy, one for recurrent high-grade T1 and the other for pubovesical fistula.
The most common any-grade TRAEs were fatigue (35%), rash (32%), pruritus (27%), hypothyroidism (22%), and increased liver enzymes (19%). Twelve patients had grade ≥3 TRAEs, the most common being elevated liver enzymes (n=5 patients) and pruritus (n=2).
Pietzak acknowledged study limitations beyond the small number of patients and non-randomized design: limited generalizability (conducted at one high-volume center with a multidisciplinary team specializing in bladder cancer), lack of mandated post-treatment biopsy to confirm response, and no assessment of patient-reported outcomes.
Source: Read Full Article
