Pembrolizumab Plus CG0070 Induces Promising Response Rates in BCG-Unresponsive NMIBC

The addition of the selective oncolytic adenovirus CG0070 to pembrolizumab (Keytruda) showed encouraging activity and safety in Bacillus Calmette-Guerin (BCG)–unresponsive non-muscle invasive bladder cancer, according to early data from the phase 2 CORE1 trial (NCT04387461) presented at the 2022 American Association for Cancer Research (AACR) Annual Meeting.¹

Treatment options for non-muscle invasive bladder cancer unresponsive to BCG are still needed since pembrolizumab is the only approved agent for this population in the past 20 years,” Ed Uchio, MD, FACS, CPI, University of California, Irvine, said during a presentation of the results. “CG 007 activity in NMIBC after BCG failure has been established in 2 past studies of monotherapy…. Early efficacy and safety data from the CORE1 study is promising for the combination of the agent plus pembrolizumab for the treatment of NMIBC, unresponsive to BCG.”

The overall complete response (CR) rate was 88.9% (n = 16/18), with CRs at 12 months occurring in 75% of patients (n = 6/8).

“On the swimmer plots, patient 11 had persistent disease at 3 months but chose not to undergo reintroduction. We had no response in patient 13. But of the 18 patients besides those 2, they all had a response,” added Uchio.

There were no grade 3 to 5 adverse events (AEs) reported in the trial. Any grade AEs included pollakiuria (n = 6), bladder spasm (n = 4), dysuria (n = 4), fatigue (n = 4), nocturia (n = 3), chills (n = 2), hematuria (n = 2), micturition urgency (n = 2), polyuria (n = 2), as well as arthralgia, autoimmune thyroiditis, blood thyroid stimulating hormone increased, device leakage, diarrhea, headache, hot flush, hyperglycemia, hypothyroidism, influenza-like illness, joint stiffness, malaise, musculoskeletal stiffness, myalgia, penile hemorrhage, pruritis, pyrexia, urinary tract infection, and urinary tract pain (n = 1 each).

Previous Studies of CG0070

The agent was previously evaluated as a monotherapy in 2 clinical trials: the phase 1 V0046 trial (NCT00109655) and the phase 2 BOND2 trial (NCT02365818).

The V-0046 trial evaluated 35 patients treated with CG0070 who demonstrated anti–bladder cancer activity with a CR rate of 46% at any time. The most common AEs were grade 1/2 bladder toxicities.²

The BOND2 trial evaluated 65 patients, with a 65% CR rate at any time and CRs maintained in 28% at 12 months.³

Overall, between both trials, over 100 patients with NMIBC were treated with approximately 800 intravesical administrations. There were no severe (grade 4 or 5) AEs. Both trials demonstrated confirmed gene expression (GM-CSF in urine) and virus replication (clear PCR evidence).

CORE1 Study

Since CG0070 appeared promising in previous studies, the investigators aimed to improve the efficacy of the agent by adding a checkpoint inhibitor, pembrolizumab.

In the open-label, single-arm phase 2 trial, with a target enrollment of 35 patients, investigators administered CG0070 weekly x 6 at a dose of 1x10¹² vp/mL, followed by weekly x 3 to 6 every 3 months for year 1; every 6 months for year 2; and in combination with pembrolizumab at 400 mg every 6 weeks through year 2.

The primary end point of the trial was CR at 12 months, and secondary end points included safety, CR at any time, progression-free survival, cystectomy-free survival, duration of response, and immune correlates on all patients.

Key entry criteria included those with NMIBC unresponsive to BCG disease; 5+2 BCG minimum prior exposure; relapse with cisplatin within the last 12 months of BCG treatment; CIS with or without Ta/T1 tissue diagnosis within 10 weeks of first study treatment; were ineligible or refused cystectomy; and could undergo mandatory bladder mapping at 12 months and central pathology review at screening and for response assessment.

Reinduction for non-progressing patients at month 3 was allowed. Patients were excluded if they previously received pembrolizumab or other anti–PD-(L)1 therapy; had ongoing, active gross hematuria; and prior adenovirus serotype 5 therapeutic exposure.

How CG0070 Works

CG0070 is a selective oncolytic adenovirus that has been engineered to preferentially replicate in tumor cells. Uchio explained that it has an E2F promoter and a GM-CSF transgene inserted into a wild-type adenovirus backbone.

“In normal cells with intact Rb tumor suppressor pathway, we don't get replication. But in bladder cancer cells, in which the Rb pathway is defective, either with hyper phosphorylation, deletion, or silencing, we get replication,” he added. “It replicates inside the tumor cells causing selective cancer cell lysis and death.”

As an oncolytic immunotherapy, CG0070 has a 2-pronged attack or mechanism of action:

• It infects the tumor cell, replicates, and kills the tumor cells and, in return, infects other tumor cells along the line.
• It has a mechanism of action stimulating the anti-tumor immune response with the GM-CSF cassette, priming and activating the T cell for tumor self-destruction.

Because therapies are limited for patients with NMIBC who are unresponsive to BCG, the researchers aimed maximally infect the oncolytic adenovirus into the bladder. To do this, the GAG layer—which impedes the binding of the viral vector to the cognate cancer cell receptor—must be removed using DMD, a mild detergent.

Therefore, to administer CG0070 into the bladder, the researchers started with a normal saline wash, followed by the DMD wash and a second saline wash, and apply the agent for 60 minutes.

Next Steps for CG0070

The agent is being evaluated as a monotherapy in the open-label, phase 3 BOND3 trial, with a target enrollment of 110 patients with NMIBC. The agent will be administered in an induction course of weekly x 6 at a dose of 1x10¹² vp/mL, followed by a second induction course of weekly x 6 at a dose of 1x10¹² vp/mL for non-responders, and a maintenance course of weekly x 3 at a dose of 1x10¹² vp/mL for completer responders. The primary end point is CR at any time.

It is also being evaluated in combination with nivolumab (Opdivo) among a target enrollment of 30 patients who are cisplatin-ineligible with MIBC (cT2-4), with or without pelvic lymphadenopathy (cN0-1), and no evidence of distant metastases prior to radical cystectomy. In the phase 1/2 CORE2 study, the regimen consists of CG0070 weekly x 6 at a dose of 1x10¹² vp/mL in combination with nivolumab every 4 weeks starting after week 1 of CG0070. End points of the study include safety, pathological response rate, and recurrence-free survival.

References

1. Li R, Steinberg G, Shah P, et al. CORE1: phase 2, single arm study of CG0070 combined with pembrolizumab in patients with non-muscle invasive bladder cancer (NMIBC) unresponsive to Bacillus Calmette-Guerin (BCG). Presented at: AACR Annual Meeting 2022; April 8-13, 2022; New Orleans, Louisiana. Abstract CT036.
2. Burke JM, Lamm DL, Meng MV, et al. A first in human phase 1 study of CG0070, a GM-CSF expressing oncolytic adenovirus, for the treatment of nonmuscle invasive bladder cancer. J Urol. 2012;188(6):2391-2397. doi:10.1016/j.juro.2012.07.097.
3. Packiam VT, Lamm DL, Barocas DA, et al. An open label, single-arm, phase II multicenter study of the safety and efficacy of CG0070 oncolytic vector regimen in patients with BCG-unresponsive non-muscle-invasive bladder cancer: interim results. Urol Oncol. 2018;36(10):440-447. doi:10.1016/j.urolonc.2017.07.005.