Jonathan Rosenberg, MD
Second-line atezolizumab (formally MPDL3280A) demonstrated significant clinical benefit in patients with metastatic urothelial carcinoma (mUC) who had a poor prognosis after progressing following platinum-based chemotherapy, according to the phase II IMvigor 210 results reported at the 2015 European Cancer Congress.
Atezolizumab restores immune response by targeting PD-L1 and can enhance T-cell priming and re-invigorate suppressed immune cells by inhibiting binding of PD-L1 to PD-1. Atezolizumab demonstrated activity previously in mUC, leading to a breakthrough designation from the FDA in 2014 for patients with mUC whose tumor expressed PD-L1. Results from the IMvigor 210 trial will be submitted under this designation to the FDA, since there is currently a high, unmet need for viable treatments in mUC.
“The status of available treatments in metastatic urothelial cancer is dismal, quite frankly,” said lead investigator Jonathan Rosenberg, MD, a medical oncologist at Memorial Sloan Kettering Cancer Center. “No single agent therapy has been observed to improve median overall survival. Vinfluinine is the only approved agent in the European Union and demonstrated an overall response rate (ORR) of just 8.6% versus 0% for best supportive care (BSC),” he continued.
IMvigor 210 enrolled an all-comer population of 316 patients, 26.4% in Europe and 73.4% in the United States and Canada. All patients had progressed during or following platinum-based chemotherapy. The patients had been heavily pretreated, with 40% of patients undergoing 2 or more prior systemic regimens in the metastatic setting and 74% of patients receiving previous cisplatin-based chemotherapy.
Data from 311 patients were evaluable; the mean patient age was 66 years, 78% were male, and 62% of patients had ECOG PS 1. The site of primary tumor was the bladder for 75% of patients with metastasis to viscera reported for 78% of patients and to the liver in 31% of patients.
Tumor tissue was prospectively assessed centrally for PD-L1 expression using the SP142 immunohistochemistry assay. PD-L1 status was assessed on tumor cells and immune cells using an SP142 antibody-based immunohistochemistry assay; however, both the patients and investigators were blinded as to PD-L1 status.
Atezolizumab was administered at 1200 mg intravenously on the first day of each 21-day cycle until no further clinical benefit was demonstrated. Median treatment duration was 12 weeks (range, 0-46 weeks). The co-primary study endpoints were ORR, as assessed by central review and the investigators, both by modified RECIST v1.1. Secondary endpoints included duration of response (DoR), progression-free survival (PFS), overall survival (OS) and safety.
When results were evaluated according to the degree of PD-L1 expression on immune cells and on all comers, significant ORR improvements across all groups were seen that increased with higher PD-L1 expression. ORR by RECIST 1.1 was 15% (P
= .0058) in all comers, 18% (P
= .0004) at the IC1/2/3 level (any PD-L1 expression), and 27% (P
= .0001) in the IC2/3 subgroup with PD-L1 expression ≥5%.
“This demonstrates significant improvement over the historical ORR of 10% in these patients,” said Rosenberg.
ORR in all comers was 12%, 18%, and 13% in patients receiving 1, 2, and 3 or more prior systemic treatment. In patients expressing PD-L1 IC2/3, ORR was 26%, 39%, and 20% with 1, 2, and 3 or more prior therapies.
An analysis of changes in target lesion by PD-L1 expression revealed a 32% lesion reduction in the IC0 subgroup, 43% in IC1, and 80% lesion reduction in patients with the highest level of PD-L1 expression (IC2/3). ORR in the respective groups was 9%, 10% and 27%.
The median DoR had not been reached at the time of cut off but demonstrated a range of 0+ to 43 months. At a median follow-up of 7 months, 92% (43 of 47) of patients demonstrated an ongoing response.
Overall, 12 patients achieved complete response (CR), and 35 achieved partial response (PR). In addition, 15 additional unconfirmed RECIST v1.1 CR/PRs were seen.
“Traditionally, complete response is very rarely observed in patients with mUC,” Rosenberg remarked.
Median PFS at a median follow-up of 24 weeks was 2.1 months across all groups. OS data were not yet mature but prolonged OS was demonstrated in patients with higher PD-L1 expression. Median OS was 7.9 month in the overall cohort, with patients having IC0/1 achieving PFS of 7.6 months compared to not reached in patients expressing PD-L1 IC2/3.
Treatment-related adverse events (AEs) of any grade occurred in 66% of patients, with 15% of patients experiencing grades 3/4 treatment-related AEs consisting most often of fatigue in 6 (2%) patients. AEs leading to treatment discontinuation were reported in 3% of patients.