Early Pembrolizumab/Bevacizumab Study Shows Promise in RCC

Shilpa Gupta, MD

The combination of pembrolizumab (Keytruda) and bevacizumab (Avastin) proved to be safe and tolerable for patients with metastatic renal cell carcinoma (mRCC), according to phase Ib/II data presented at the 2018 ASCO Annual Meeting.

The phase Ib part of the trial evaluated 13 patients who previously progressed on at least 1 line of therapy. Patients received 200 mg of pembrolizumab along with either 10 or 15 mg/kg of bevacizumab every 3 weeks. Investigators reported no dose-limiting toxicity levels.¹

The phase II portion of the trial, said Shilpa Gupta, MD, tested the standard dose of pembrolizumab and bevacizumab in 48 patients with treatment-naïve mRCC, with the primary endpoint being objective response rate (ORR). Results showed that the ORR was 60.9% (28 partial responses) among 46 evaluable patients, which she said far surpassed expectations.

“We will probably have to do another larger study and compare it with the other ongoing combinations,” said Gupta, who is an author on the study. “The field is changing so rapidly. However, this study sends us in the right direction.”

In bladder cancer, Gupta led a study focused on genomic characterization of variant histology in muscle invasive bladder cancer (MIBC) to predict response to chemotherapy.² The trial, which evaluated 1 patient with sarcomatoid bladder cancer and 1 with small cell bladder cancer, had data showing that the genomic profiles provide information for predictors of response and treatment selection.

OncLive: Can you provide some background on the phase Ib/II studies of pembrolizumab and bevacizumab?

In an interview with OncLive, Gupta, an assistant professor of Medicine in the Hematology/Oncology and Transplantation Division at the University of Minnesota, discussed the tolerability and safety of pembrolizumab and bevacizumab in RCC, as well as the future of bladder cancer treatment in the neoadjuvant setting.Gupta: This is a study dealing with patients who have mRCC. The phase Ib [portion included] patients who had progressed on at least 1 previous therapy. There were 13 patients enrolled on this and the combination was very safe and very well tolerated. The phase II portion had [48] treatment-naïve patients with mRCC enrolled. They got [first-line treatment with] pembrolizumab and bevacizumab at standard doses and the primary endpoint was ORR.

Why is there such intriguing synergy with this combination?

There are a few different targeted therapy and immunotherapy combinations being studied in the RCC paradigm. Is there anything about pembrolizumab and bevacizumab that stands out from the rest?

ill we ever see a point when we are looking at triplet therapy for mRCC?

You are also an author on a MIBC study at the 2018 ASCO Annual Meeting. Can you give an overview of that trial?

There was over a 60% ORR, beating by far our hypothesized responses. This is a very promising combination that makes sense scientifically, and patients tolerated the treatment very well. There was no distinction between high- or low-risk in this study. Our ORR was seen across all [subgroups]. We are very excited about these data.The histogenesis is the backbone of kidney cancer. Using bevacizumab leads to more trafficking; it controls the trafficking of the tumor cells. The synergy between these agents and checkpoint inhibitors, as we have seen in other types and kidney cancer, is the rationale for combining these. The tolerability is also not overlapping.That is difficult to say because, like you said, there are a lot of trials going on. For example, bevacizumab and atezolizumab (Tecentriq) is another promising combination. However, that is only for patients with a 1% or higher PD-L1 expression. Our study did not require that; therefore, that makes it stand apart. Also, it tells us that PD-L1 expression does not matter. Our ORR across the board, regardless of PD-L1 expression, is really impressive. That is a highlight of this study.W We need some preliminary safety data, because there are some data coming out from many tumor types that combinations may be more effective but may not be that safe. With ipilimumab (Yervoy) and nivolumab (Opdivo), we have seen higher efficacy, but toxicity levels were not minimal. To that, adding another agent may compound the issue, [but] we need some phase I trials to see that.The molecular subtypes in MIBC, based on the gene expression profiling, have led to our understanding of how these different subtypes can affect the response to neoadjuvant chemotherapy. We have an abstract at [this meeting] where we are looking specifically at variant histologies in MIBC. Therefore, the 2 cases we highlighted are 1 of sarcomatoid bladder cancer and 1 of small cell bladder cancer. These are rare types, but we often see them in clinical practice.

The first case received standard chemotherapy. The patient blew through chemotherapy at rapid progression and, within 4 weeks, developed widespread metastases. We started her on pembrolizumab and, remarkably, she got a complete response and remains in complete remission. We looked at her molecular subtyping from the bladder biopsy and it showed a high epithelial-to-mesenchymal transition, which would explain lack of response to chemotherapy.

We also did next-generation sequencing on her metastatic sites, and that showed that she had a high PD-L1 expression. We need to keep doing this to decide what neoadjuvant therapy is best for each patient. For example, in this patient, she was likely a good candidate for immunotherapy.

Are there similar types of analyses that you are interested in pursuing?

How widespread are genomic classifiers being used right now?

Our second patient had a response to chemotherapy later on, so this series helps us understand how the various histologies in MIBC perform. It's exciting to see where our field will move to and decide how to personalize presurgical approaches.There have been analyses on the small cell bladder cancer cases. While it is not the standard to use these genomic classifiers, they certainly help our understanding and complement what we know. However, we need larger prospective trials like the ones we are doing.Their usage varies across the providers. Some institutions are routinely ordering this. Sometimes the urologists send us the results and we say, "Wow, we don't think this patient really needs chemotherapy." Therefore, we have a discussion that it is a guidance, but we should still offer neoadjuvant chemotherapy when we can.

The more we do these tests and see these results, then we can come with a better understanding. At our institution, our urologists order a lot of these [assays], and across the country they are being ordered a lot. However, I do not know how many people are solely basing their decisions on them.

References

1. Dudek AZ, Liu L, Gupta S, et al. Phase Ib and phase II studies of pembrolizumab (P) with bevacizumab (B) for the treatment of metastatic renal cell carcinoma (RCC): BTCRC-GU14-003. J Clin Oncol. 2018;36 (suppl; abstr 4558).
2. Gupta S, Konety BR, Davicioni E, Gibb E. Genomic characterization of variant histology muscle invasive bladder cancer (MIBC) to predict response to therapy. J Clin Oncol. 2018;36 (suppl; abstr e24290).