AACR 2016

5 year survival rates for nivolumab-treated metastatic melanoma patients much higher than historical rates

Dr Stephen Hodi - Dana-Farber Cancer Institute, Boston, USA

Thank you, and thank you everyone for the opportunity to speak to you this afternoon. So on behalf of my co-authors I’m very pleased to present to you an update on our data from the nivolumab monotherapy phase I trial. Here are my disclosures.

As part of the background information, nivolumab, or nivo, is an immune checkpoint inhibitor that blocks the programme death-1 signalling that enhances anti-tumour immune responses. Nivolumab is approved for first line treatment of advanced melanoma, both as a single agent and in combination with ipilimumab, or ipi, anti-CTLA4. The efficacy and safety of nivolumab monotherapy in advanced malignancies was validated in a phase I dose escalation study. In the current report we are demonstrating the overall survival rates for five years of follow-up for advanced melanoma patients in the so-called study 209003. This represents the longest survival follow up to date for any anti-PD-1 agent in any specific disease.

This is to remind you of the study schema. Patients were eligible with advanced melanoma, a total of 107 patients were treated. Patients were typically heavily pre-treated, they could have received from one to five lines of prior systemic therapies. Included in this is the dose escalation from 0.1mg/kg up to 10mg/kg and the treatment duration could last for up to 96 weeks. The key eligibility for this patient population included melanoma from any primary site but it could have included also ocular and mucosal melanomas.  They could have had treated and stable disease from brain metastases, measurable disease by standard criteria, a good performance status and good shape from 0, 1 or 2 ECOG performance status and had at least one prior systemic therapy. They could not have had prior anti-CTLA-4 ipilimumab or another anti-PD-1 or anti-PD-L1 antibodies. The primary objective was safety and tolerability, second is efficacy and dose response and the study was eventually amended to collect overall survival data long term and as an exploratory analysis to examine outcomes of patients who were re-treated with nivolumab therapy, which we’ll discuss in a moment.

So the current dataset that we’re presenting today shows an analysis that was completed in October of 2015. We have an overall survival with a minimum follow-up of 45 months. We’ve updated safety data and updates on the patients who are still in response, including those patients who underwent retreatment with nivolumab monotherapy.

This represents the table of the key baseline characteristics: the median age of 61 years. There was a predominance of male versus females who were enrolled. Most patients had very good performance status, a majority of 0 and 1 versus 2. Two-thirds of patients had more than two prior systemic therapies and most of that was prior immune therapy, including 46% of patients who had been previously treated with interleukin-2 immune therapy. A majority of patients had visceral metastases and about a third of patients had elevations in lactate dehydrogenase, LDH, which is typically a poor prognostic sign for patients with a metastatic melanoma.

So this is the survival curve for all patients treated. So 34% of patients survived five years and, of note, there does seem to be around 48-54 a beginning of a plateauing in the survival curve that now appears to be durable, lasting months to years. This curve overlapping this is the nivolumab group of patients who were treated with 3mg/kg and this is just to note that this is the dose that went forward in phase III trials for registrational aspects and shows that it’s pretty similar in this aspect in that both the nivolumab patients who received the 3mg/kg dose and all patients seemed to exhibit this plateauing here at this so-called that’s lasting many months to years, about 34%, about a third of patients having this long-term survival. The median overall survival is being reported here as 17.3 months.

Now, interestingly there were some patients who were re-treated with nivolumab and here we demonstrate some of the data for five of the patients. Patients were permitted to be re-treated with nivolumab monotherapy under specific conditions: those patients who achieved disease control and subsequently went on to develop progressive disease; who do not experience a dose limiting toxicity with the previous treatment; who do not reach the one year follow-up without progressive disease or any patient who had already undergone re-treatment. So five patients were re-treated in the same dose as they were originally assigned to after being off treatment for nivolumab for more than 100 days. So this are the so-called Spavey plots or spider plots, so zero is the total tumour burden the patient begins with at re-treatment and then this is a percentage of going up or down, so going up things are getting bigger and going down things are getting smaller. You can tell in all five patients disease control was obtained again and this disease control was again durable. So this one patient continues on treatment here, with a circle, interestingly this patient here had an adrenal metastasis that was removed, one site of disease was removed, making them basically no more evidence of disease and this has actually led to a long-term durable benefit, suggesting the importance of some of the memory aspects of giving checkpoint blockade and its adaptability long-term to help patients.

These are the most common related adverse events to nivolumab therapy, including fatigue in about a third, rash, this is any grade and this is the high grade events. Essentially there are no new safety signals with long term follow-up compared to the original data that has been presented previously and, importantly, no study drug related deaths.

Here is an attempt to put these data into even a greater perspective. So the green here represents the survival curve for patients who receive anti-CTLA4 ipilimumab therapy. It’s over 1,800 patients with up to 5-10 years of follow-up showing that around 2½ -3 years this inflection point where about 22% of patients have long-term survival. Here is the current study that we’re discussing of 107 patients long-term nivolumab therapy with up to 5 years of therapy showing a little bit better here this inflection, the flattening of the curve happening somewhere around 3 years here with a durability of survival there. What this curve represents is actually a trial, the so-called 6 trial of nivolumab monotherapy. It’s heavily censored here but we are awaiting a read-out of this trial. But it is a different patient population where this was a front-line patient population where what I presented to you here are patients who have been previous treated, many of them heavily pre-treated. So we await the maturity of this dataset here for patients in the front-line setting.

But what it does demonstrate is the importance of the durability of clinical benefit to patients now we’re measuring in terms of years. As well as the memory aspect of how the immunologic memory can translate into benefit here with ipilimumab, now with nivolumab, now asking the question front line, which we need more data, and many other studies with combinations that we’ll also hear about later this afternoon as well as future studies.

So, to conclude, at a minimum follow-up of 45 months nivolumab monotherapy resulted in a five year overall survival rate of 34% in heavily pre-treated patients with advanced melanoma. The overall survival rates appeared to plateau at about 48 months; the updated analysis of re-treated patients showed that disease control was maintained in those patients. Nivolumab continues to be safe and tolerable with no deaths and no new safety signals and these data represent the longest survival follow up of patients who received anti-PD-1 therapy in a clinical study and suggests durable long-term survival with nivolumab monotherapy. Thank you very much for your attention.