We then talked about the second-line setting, and this is where the drugs have been the most well investigated—in the post-platinum setting. That could be in any setting: neoadjuvant therapy, adjuvant therapy, or for metastatic disease that has already been treated with platinum therapy. There have been 2 phase III studies performed in that setting and a bunch of phase II studies. There are 5 agents approved in that setting: pembrolizumab, atezolizumab, durvalumab, avelumab, and nivolumab. The pembrolizumab and atezolizumab data are the most robust because they are from phase III studies. The response rates and the toxicity rates were very similar to what I already said with a 20% to 25% response rate and the toxicity looks fairly good.
We did sort of drill in on the atezolizumab phase III study IMvigor211, which was presented a few months ago. It is a confusing study for a lot of folks, because the study was called negative if you read the press release. However, the way the study was designed led to that being negative. If you look at the overall survival (OS) of atezolizumab compared with chemotherapy, which is the comparator, results showed that in all comers, there was a statistically significantly benefit with atezolizumab. The curves separated as you might expect.
The way the study was designed, though, is that they were initially looking at just the PD-L1–high subset. In that group, there was not a significant difference between atezolizumab and chemotherapy—rather, it was in all comers. Therefore, if the study was designed a little differently, then that would have been a positive study. Although the study was called negative, the jury is still a little bit out. The drug clearly has activity based on a lot of data that have already been presented. In the frontline setting, pembrolizumab has a clear survival benefit. It is a level 1 recommendation.
The other agents are all in different trials and in different settings in bladder cancer. The studies of frontline immunotherapy are either as monotherapy or in combination with chemotherapy. Then, there are a number of studies earlier on, such as combinations with BCG for non-muscle invasive disease or after BCG for patients who have not had their bladders removed but maybe need cystectomy. There are adjuvant studies going on after removal of the bladder to see if we can cure people who would otherwise relapse, as well as combinations with radiation therapy.
We switched and talked about targeted therapy. I focused on 3 major developments, the first being FGFR. That is mutated or overexpressed…in about 20% of muscle invasive tumors; that might be higher in metastatic tumors. Essentially, there are antibodies being developed that inhibit the receptor. We reviewed the data that was presented from the FGFR inhibitor BGJ398, which showed some durable long-term responses. That is really exciting for targeted therapy in bladder cancer, because there has not been many studies of targeted therapy that have shown durable responses like that.
There are a number of other compounds in trials, so that is an exciting next step because it is independent of the immune system in a way; it sort of offers another opportunity for bladder cancer patients to benefit.
There was recently a phase III study presented of ramucirumab (Cyramza) and there was a modest benefit seen in the second-line setting in post-platinum metastatic patients when given together with a taxane. The benefit was modest—a 1.5-month difference in progression-free survival (PFS). Is that clinically meaningful? Thinking about the future, does that mean we will use ramucirumab in combination with chemotherapy, or should these be added to the armamentarium for adding it to a PD-1 inhibitor or to other contexts? We are going to see trials looking at that to figure out how to best use this. With bevacizumab (Avastin), there are studies we are waiting on data for.
We mentioned targeting ERBB2, which is mutated or overexpressed in bladder cancer in 10% or 15% of patients. Unfortunately, targeting it with trastuzumab (Herceptin) or lapatinib (Tykerb) in phase II/III studies has not really panned out. There hasn’t been a tremendous benefit observed, which could be due to the study design, but it may also be that it’s just not as important as a driver of bladder cancer.
Putting that all together, it is an exciting time in bladder cancer—between immunotherapy and targeted therapies. The treatment landscape is going to change. If we come back 5 years from now, it’s going to look very different. It is a great time to be treating these patients and it is great to be able to offer so much for them.
In the second-line setting, how do you decide which of the 5 checkpoint inhibitors to use?
Great question. The overarching answer is we don’t really know. The cleanest data, I should say, is for pembrolizumab from the KEYNOTE-045 study because there was an OS benefit. It was clear that it was independent of PD-L1 expression, and if you look at NCCN, that’s essentially the category 1 recommendation. There is a lot of rationale still to give atezolizumab in this setting if you look at the intent-to-treat population; it was a positive study. It is just a challenge.