Atezolizumab Shows Promise of PD-1/PD-L1 Blockade in Metastatic Bladder Cancer

Immune checkpoint blockade has transformed the treatment paradigms of multiple advanced solid tumors-and bladder cancer is no exception.

Matthew D. Galsky, MD

Associate Professor of Medicine

Icahn School of Medicine at Mount Sinai

Director of Genitourinary Medical Oncology, Tisch Cancer Institute

Metastatic urothelial cancer of the bladder is a relatively chemotherapy sensitive malignancy. However, response durations are generally short-lived with standard platinum-based combination regimens and the vast majority of patients ultimately succumb to the disease. For patients progressing on first-line chemotherapy, options are limited with no treatments approved by regulatory authorities for use in this setting in the United States. In fact, there have been no new treatments approved by the FDA for the treatment of bladder cancer in over 30 years.1

Immune checkpoint blockade has transformed, or is on the verge of transforming, the treatment paradigms of multiple advanced solid tumors and bladder cancer is no exception. Expansion cohorts of phase I studies with the anti—PD-L1 antibody, atezolizumab, and the anti–PD-1 antibody, pembrolizumab, have demonstrated durable responses in a subset of heavily pretreated patients with metastatic bladder cancer with some correlation observed between higher expression of PD-L1 in either tumor or tumor infiltrating cells (depending on the particular study) and increased likelihood of response.2,3

Preliminary results of the IMvigor 210 study, the first large phase II study exploring immune checkpoint blockade in metastatic bladder cancer, were recently presented at the 2015 European Cancer Congress.4IMvigor 210 enrolled 316 patients with metastatic bladder (or with urothelial cancers originating elsewhere in the genitourinary tract) who received treatment with atezolizumab at 1200 mg (IV) every 3 weeks until loss of clinical benefit. All patients had tumor tissue submitted prior to enrollment for evaluation of PD-L1 expression in tumor infiltrating cells using the Ventana PDL1 (SP142) CDx Assay; however, PD-L1 expression was not an eligibility criterion for enrollment.

The primary endpoint of the study was objective response rate (ORR) per RECIST v1.1 and the primary endpoint was met if the null hypothesis, an ORR of ≤10%, was rejected at a significance level of 5%. This threshold of activity was selected based on historical studies evaluating a variety of systemic therapies in platinum-resistant metastatic bladder cancer.

The study met its primary endpoint demonstrating an ORR of 27% (95% CI, 19%-37%) in patients with tumors harboring 2-3+ PD-L1 expression in tumor infiltrating cells and 15% (95% CI, 11%-20%) in all comers, regardless of PD-L1 expression. Importantly, responses were durable and the median duration of response was not reached in any subgroup (according to PD-L1 expression) at the time of the data cutoff. Responses were seen across clinical patient subgroups, including those with visceral disease (including liver metastases) and those who had received ≥3 prior systemic treatment regimens.

Treatment was generally well tolerated, with grade 3/4 treatment-related adverse events in 15% of patients and grade 3/4 immune-related adverse events in 4% of patients.

Importantly, IMvigor also enrolled a second cohort of patients, chemotherapy-naïve cisplatin-ineligible patients with metastatic bladder cancer.5The results in this group of patients, also representing a major unmet need in the field, have not yet been reported.

Multiple ongoing trials will define the role of PD-L1 and PD-1 blockade in patients with metastatic urothelial cancer of the bladder, as well as in other disease states of bladder cancer, including the perioperative setting and in the treatment of non—muscle-invasive disease. These trials include additional large phase II and randomized phase III trials poised to secure regulatory approval for these agents in the near term.

The results of the IMvigor trial, and the growing experience with other PD-L1 and PD-1 antibodies in bladder cancer, raise several critical questions. Indeed, these questions are not unique to bladder cancer but are questions common to the application of this novel drug class across malignancies: Will PD-L1­—based predictive biomarker assays ever be ready for “prime time,” given the variety of antibodies and scoring systems being explored, particularly in the context of responses observed in patients without biomarker expression and the dynamic nature of this biomarker? How can we increase the proportion of patients benefitting from these therapies? Can treatment ever be stopped in patients with durable disease control?

Clearly, we have work to do and the excitement surrounding PD-L1 and PD-1 blockade in metastatic bladder cancer has little to do with the overall proportion of patients responding and everything to do with the durability of disease control in those patients who do respond, accompanied by the favorable tolerability and safety of these drugs. Seeing patients still responding to PD-L1 and PD-1 blockade 1…1.5…2+ years out from initial treatment in the “salvage” setting is truly game-changing in this disease and provides an entirely new platform on which to build for the future.

References

  1. Galsky MD, Domingo-Domenech J. Advances in the management of muscle-invasive bladder cancer through risk prediction, risk communication, and novel treatment approaches. Clin Adv Hematol Oncol. 2013;11(2):86-92.
  2. Plimack ER, Bellmunt J, Gupta S, et al. Pembrolizumab (MK-3475) for advanced urothelial cancer: Updated results and biomarker analysis from KEYNOTE-012. J Clin Oncol 33, 2015 (suppl; abstr 4502).
  3. Powles T, Eder JP, Fine GD, et al. MPDL3280A (anti-PD-L1) treatment leads to clinical activity in metastatic bladder cancer. Nature. 2014;515(7528):558-562.
  4. Rosenberg J, Petrylak D, Abidoye O, et al. Atezolizumab in patients (pts) with locally-advanced or metastatic urothelial carcinoma (mUC): Results from a pivotal multicenter phase II study (IMvigor 210). Presented at: 2015 European Cancer Congress; September 25-29; Vienna, Austria. Abstract 21LBA.
  5. Galsky MD, Hahn NM, Rosenberg J, et al. A consensus definition of patients with metastatic urothelial carcinoma who are unfit for cisplatin-based chemotherapy. Lancet Oncol. 2011;12(3):211-214.