ESMO 2016

Adjuvant immunotherapy improves melanoma outcomes

Prof Alexander Eggermont - Institut Gustave Roussy, Villejuif, France

Good morning. The trial that I will discuss with you is the EORTC 18071 trial, this is a phase III randomised trial that investigates the question whether ipilimumab in the adjuvant setting compared to placebo after complete resection of stage 3 melanoma, this is lymph node positive melanoma, brings a benefit to patients. What we are reporting here is final overall survival results from this double blind phase III trial.

This went just online at the website of The New England Journal of Medicine under the title ‘Prolonged survival in stage 3 melanoma with ipilimumab adjuvant therapy’ so you know what the message is going to be.

This is the trial design. We randomised 951 patients with lymph node positive, high risk for relapse, stage 3 melanoma patients into either receiving ipilimumab induction four doses, three weeks apart over the first three months, and then maintenance therapy every three months at a dose of 10mg/kg up to three years or receive a placebo intravenous treatment according to the same schedule. Patients would therefore come off treatment for reasons, either relapse or come off treatment because of side effects or come off treatment because of having completed the three years.

Stratification factors are stage 3a versus 3b versus 3c and then, depending on the number of positive nodes, 1-3 nodes or 4 or more nodes positive, because this is the most important prognostic factor in stage 3 melanoma. And we stratified by region – the trial was conducted in Europe, in the US and in Australia.

I immediately go to primary endpoint. Primary endpoint is relapse free survival but primary endpoint analysis was already mature last year and we published it in Lancet Oncology. But this is now at 5.3 years median follow-up, the publication in Lancet Oncology is with 2.7 years median follow-up. What it shows is exactly the same results as at 2.7 years median follow-up. So it shows a hazard ratio of 0.76, that means a 24% reduction of risk for relapse or death. You see the p-value is 0.008 and you see the absolute difference in terms of median relapse free survival which is 10.5 months. It’s the endpoint that we can now report on as new information and it’s the impact on overall survival. The impact on overall survival has a hazard ratio of 72%, of 0.72, look at the upper right panel, that means there is a 28% reduction of risk of death. It also gives you the precise numbers. So 162 patients in the ipilimumab arm died out of the 475 patients compared to 52 patients more died in the placebo arm, that means 214 over 476. This is just a slide that you need to understand and then you’re fine. What it also means is at five years in terms of what is the death rate difference in absolute numbers it’s 11%. So the survival rate is 65% for ipilimumab, it’s 54% for placebo.

Why do I have a shaded area on this slide at seven years? It is because you actually cannot look at that right end of the curve, there are just a couple of patients at risk. So that’s not solid observation. Here at five years you have 200 and 178 patients at risk so you know exactly where we are. OK?

Now, of course, it comes at a price; it comes at a price in terms of side effects and toxicity. Here you see listed in all grades, grade 3, 4 and 5 toxicities, in the ipilimumab column and grade 3, 4 in the placebo column what happens. Now, we focus here on immune related adverse events because these are by far and large the majority of the adverse events associated with ipilimumab therapy. You have five major blocks of adverse events – dermatologic, gastrointestinal, endocrine, hepatic or neurologic - but it’s most telling to look at grade 3/4 events because these are the events that make the patients come off treatment. So dermatologic – it’s a rash and it may be inconvenient but it’s not so important, but gastrointestinal is very important because gastrointestinal has diarrhoea and colitis. It basically is an autoimmune inflammatory process that’s unleashed by ipilimumab. When you look at grade 3/4 events you see that overall it’s in about 16% of patients. The colitis is the most important one because it can also be associated with perforation and actually in three patients it led to death, that’s grade 5 toxicity.

Then endocrine adverse events, immune related adverse events are mainly hypothyroiditis, so a lower functioning of the thyroid gland, but most importantly hypophysitis, it means that your pituitary gland stops functioning. That happens in about 4.5% of patients; the problem with that is that you most likely will have to have lifelong hormonal replacement therapy. Then hepatic immune related adverse events, it’s autoimmune hepatitis, and you see that in total in about 16% of patients at the grade 3/4 level. In terms of neurologic immune related adverse events there is a syndrome called Guillain–Barré. The Guillain–Barré actually can put you… it’s a peripheral viral infection that is causing paralysis and you may even be put on a mechanical ventilator. One patient was in that category but died of septic shock, multi-organ failure, and one other patient had an autoimmune myocarditis and also died of drug-related adverse events. So grade 5 events is 1.1% in this trial. Overall immune related adverse events grade 3/4 in 42% of patients compared to 2% in the placebo arm.

What does it mean when you compare the outcome of this trial to interferon adjuvant therapy? What it shows is that adjuvant interferon therapy in 2,644 patients, so there is a non-significant impact on microscopic lymph node metastatic disease, the hazard ratio overall is 0.81. There is absolutely no impact on overall survival in patients with palpable node disease which you see in these two overlapping curves, hazard ratio is 1.0, it means zero benefit of adjuvant interferon therapy. In this ipilimumab trial for patients with microscopic disease the blue and red curve are significantly apart from one another and for the first time now we see also in palpable nodal disease that treatment does better than placebo at a hazard ratio of 0.80. So it performs better in all categories than what is until now adjuvant therapy.

The summary of the conclusions is ipilimumab provided prolonged overall survival, distant metastasis free survival and relapse free survival. It was consistent across all survival endpoints and in absolute terms the consistency is 11% for overall survival, 11% for distant metastasis free survival and 11% for relapse free survival so it’s very consistent. The safety results remain with the conclusion that ipilimumab is not an easy drug to handle, my recommendation is to keep it in centres, and that we have an increased adverse event rate with this dose of ipilimumab. But it does now present an important option for patients to be treated in the adjuvant setting. Thank you.