In patients with bacillus Calmette-Guérun–unresponsive non-muscle invasive bladder cancer who are being treated with nadofaragene firadenovec, elevated levels of antibody titers may be able to predict efficacy.
Increased levels of antibody titers and fold changes could be predictive of nadofaragene firadenovec efficacy in patients with bacillus Calmette-Guérun (BCG)–unresponsive non-muscle invasive bladder cancer (NMIBC), according to Anirban P. Mitra, MD, PhD, during a presentation on a phase 3 study (NCT02773849) at the 2021 American Urological Association Annual Meeting.1
Specifically, investigators hypothesized that anti-adenoviral antibody levels and fold change metrics could be as novel predictors of durable response to nadofaragene firadenovec. Based on response to the intravesical gene therapy, patients with peak titer levels that were greater than 800 and peak fold changes greater than 8 were determined to be favorable for response. Patients who did not meet both criteria were designated as unfavorable.
At month 12 following the first treatment with nadofaragene firadenovec, 47 patients were responsive to treatment and a total of 44 patients were nonresponders. On-treatment peak titer levels greater than 800 were reached in 89.4% of responders compared with 59.1% of nonresponders (P = .001). Regarding peak fold change, 51.1% of responders had a peak fold change of at least 8 compared with 27.3% of nonresponders (P = .020).
Further, the combination of markers was noted in 46.8% and 18.2% of responders and nonresponders, respectively (P = .004).1
“[Peak titer levels] were associated with the highest sensitivity, accuracy, and negative predictive value,” said Mita, a urologic oncology fellow at The University of Texas MD Anderson Cancer Center in Houston. “In addition, when we looked at peak fold changes with a cutoff of 8 this was also able to significantly distinguish between responders and nonresponders such that we then classified patients as ‘favorable’ if they had a combination of peak titer greater than 800 and a peak fold change greater than 8.”
Mita added that the marker combination was associated with the highest sensitivity and positive predictive value.
To conduct this analysis, investigators performed anti-adenoviral antibody assessment using a quantitative ELISA 1 to 24 hours prior to treatment at baseline and before each dose at 3, 6, 9, and 12 months or at the study-withdrawal visit. The assay assessed for both raw titers as well as fold change, which was calculated as the ratio of on-treatment antibody levels compared with baseline levels. The final treatment response was assessed at the trial’s primary end point of 12 months following the first dose or at the time of high-grade recurrence.1
Patients enrolled to the trial received a single intravesical 75-mL dose of nadofaragene firadenovec (3 × 1011 viral particles/mL) over 1 hour.1 The replication-deficient recombinant adenovirus is designed to deliver interferon–α-2b (INFα-2b) to the bladder epithelium. The therapy mimics viral infection, resulting in IFNα-2b production and subsequent tumor regression offering a novel treatment for patients whose tumors do not respond or recur after BCG.2,3
In the absence of high-grade recurrence, patients received repeat dosing via catheter at 3 months, 6 months, 9 months, and 12 months; additional dosing was permitted at investigator’s discretion. The primary end point was complete response at any time. A total of 157 patients were enrolled and treated in the study and nonresponders were reported at 6 months (n = 21), 9 months (n = 8), and 12 months (n = 15).
The efficacy population included 151 patients and a complete response was observed in 59.6% (95% CI, 51.3%-67.5%) of patients at 3 months. The duration of response or high-grade recurrence-free survival was 7.31 months (range, 5.68-11.93).3
The optimal titer threshold was determined based on the assessment of peak titers and peak fold change. For both responders and nonresponders, peak titers were achieved for most patients approximately 6 months after initiation of treatment; peak fold changes were also observed to occur by month 6. Therefore, investigators selected the optimal antibody titer level threshold to be 200 at baseline and 800 post treatment when maximizing the area under the receiver operating characteristic curve.1
Mitra noted that baseline level (200) was not associated with a difference between responders and nonresponders; “however, a posttreatment cutoff of 800 was able to significantly distinguish between responders and nonresponders at 3 months and importantly also at peak titer levels.”
Most crossovers between low- and high-titer subgroups occurred in the first 6 months of treatment. After this time point the highest proportion of responders continued to have posttreatment titers greater than 800, further supporting the use of peak titers as a marker to predict durable response.
Finally, Mita presented an exploratory analysis of baseline titers levels as predictive markers for posttreatment titer level. Despite not being associated with durable response, low baseline titers were associated with low posttreatment titers (≤ 800). High posttreatment titers levels (> 800) were significantly associated with patients who had high baseline titer levels.
Specifically, for those with high posttreatment levels, the median baseline antibody level at 3 months was 400 (interquartile range [IQR], 200-1600) compared with 200 (IQR, 125-400) for those with low posttreatment titers (P < .001).
For those who had high posttreatment titers at peak, the median baseline antibody level was 400 (IQR, 200-800) vs 200 (IQR, 1-400) for those with low posttreatment titers (P < .001).