What I’ll be presenting today is dose expansion data from study 10 focussing on patients with metastatic urothelial cancer who are refractory to platinum-based chemotherapy. Here, where we are testing the combination of durvalumab and tremelimumab, durvalumab is an anti-PD-L1 antibody, tremelimumab is an anti-CTLA-4 antibody. Essentially this study was designed to ask the question does the combination therapy of dual immune checkpoint inhibition targeting both PD-L1 and CTLA-4 does that lead to enhanced responses in this patient population.
What did you find?
In nearly 170 patients that were treated first I always begin by describing the baseline patient population that was enrolled which was highly representative of the patient population that has bladder cancer. The median age was 66, it was male predominant. About 65% of patients had had at least one prior line of chemotherapy so really it was truly a second line population that was represented in this trial. 81% of patients had visceral metastatic disease including 32% who had liver metastases. So this was a poor prognostic subgroup of patients who were enrolled in this trial. It had a median follow-up of 11.6 months; the objective response rate was 20.8%. There was also an analysis of responses based on PD-L1 status high or low. In this trial PD-L1 expression was assessed using the SP263 assay on the VENTANA platform and measured PD-L1 expression as high as being greater than 25% expression in either tumour cells or immune cells. So for patients who had had higher levels of PD-L1 expression the response rate was actually 29%, however, we also observed responses in patients who were low levels for PD-L1 expression at actually 15%.
What about the safety of this combination?
The safety of this combination was well tolerated. The treatment related adverse events that we would expect for a CTLA-4 based combination is more or less what we observed in this trial. About 28% of patients experienced a treatment related grade 3 or 4 adverse event which is similar to what we would expect. But, importantly, many of the immune-related toxicities that clinicians become very concerned about such as diarrhoea, colitis, those types of events were relatively infrequent at roughly 3% of higher grade toxicity. So overall this therapy was well tolerated and what we would expect for this combination.
What was found in the PD-L1 negative segment?
One of the interesting aspects of this trial was to really test if the patients who received combination immunotherapy if their subgroups had particular benefit. In particular are there subgroups that typically would not necessarily benefit as well from single agent therapy? So, for instance, in this trial the PD-L1 negative patients had a response rate of 15%. If we look back at study 1108 which tested single agent durvalumab therapy the PD-L1 negative patients had a 5% response rate. Obviously these are two different trials, two different sets of patients but it begins to actually lend credence to the idea that dual immune checkpoint blockade may be justified in patients who have low levels of PD-L1 expression to warrant this therapy because response rates are otherwise so low in the patients who receive monotherapy.
What’s the next step?
The next step, obviously, is to test this combination in a much larger study and in a randomised trial. One very important study that is aimed to address this question is the DANUBE study which is a randomised phase III trial of front-line durvalumab plus tremelimumab versus chemotherapy and has completed enrolment and we should expect data soon. It’s ultimately designed to ask the question does dual immune checkpoint blockade improve survival in the first line setting.
