I’m going to discuss the results of PD1 blockade with pembrolizumab in patients with classical Hodgkin lymphoma. All these patients have previously failed brentuximab vedotin.
As you just learned, classical Hodgkin lymphoma has a paucity of Reed-Sternberg cells which is the neoplastic cell in Hodgkin lymphoma but there’s a brisk mixed inflammatory background. Many of these cells are T cells and they express PD-1 and PD-1 is inhibitory in classical Hodgkin lymphoma.
It is very interesting, the story about 9p24, and its resultant overexpression of both PD-L1 and PD-L2. In addition, half of Hodgkin lymphoma patients have been exposed to EBV. That in itself is probably not prognostic but it does upregulate PD-L1 and PD-L2. In fact, it’s unclear to me if you even need to stain Reed-Sternberg cells for PD-L1 and PD-L2 because it’s almost uniformly seen.
So pembrolizumab is a humanised monoclonal IgG4 antibody against PD-1, there’s a high affinity for the receptor. It’s dual blockade of both PD-L1 and PD-L2. PK dosing is evolving but it can be given every 2-3 weeks. It has demonstrated a wide variety of activity in solid tumour, as you know this agent was approved in melanoma this year. The doses were escalated up to 10mg/kg with very little side effects which was the reason that you can see in our study we used 10mg/kg. That will not be the dose that will be moved forward, it will be a dose that’s analogous to nivolumab.
This study included patients with classical Hodgkin lymphoma. They were either transplant ineligible because of primary refractory or progressive disease, that’s a very poor patient population, or the patients failed a stem cell transplant. All the patients failed brentuximab vedotin. Hodgkin lymphoma patients usually are in good shape, they have a good performance status. All of the checkpoint inhibitors have been associated with some mild autoimmune phenomena and endocrinopathies and patients with interstitial lung disease were excluded specifically. Patients with a history of gliotoxicity or radiation pneumonitis probably should not get a checkpoint inhibitor.
Patients received pembrolizumab; those patients who had evidence of responsive disease could stay on study until they achieved no more clinical benefit. They could be taken off study if necessary for a transplant. Patients with progressive disease were discontinued.
Baseline characteristics: the follow-up of this study is somewhat shorter than nivolumab. 29 patients will be discussed today, the median age is 32; remember the median age of Hodgkin lymphoma in the United States is 28. Patients had a good performance status. The number of prior therapies were greater than five in more than half. All the patients failed brentuximab vedotin or brentuximab vedotin failed the patient, whichever you believe is the right terminology. The majority of patients failed a stem cell transplant.
Concerning adverse events an extremely well tolerated treatment, a few grade 3 events. Actually the patient who had hypoxia did not have pneumonitis. One patient had axillary pain that was severe at the tumour site, probably related to tumour inflammation, but very few events.
Here’s the waterfall plot which is really quite good. Almost all the patients had some evidence of tumour shrinkage. The complete responses are in red, partial responses in light blue, stable disease is in purple. This is anti-tumour activity based upon investigator review. These are the cohorts, kind of divided between transplant ineligible and transplant eligible. The overall response rate is 66%; 6 patients had a complete response based upon PET imaging, a number of patients had stable disease. I actually think stable disease is quite important with these checkpoint inhibitors, I’ll show you that in the next slide. But the clinical benefit rate is 86%.
The reason I think it’s important is that some of the patients who have been on the longest treatment actually have stable disease. In fact, the top person who has stable disease had tumour shrinkage by about a third, and this patient continues to get therapy. The median duration of response has not been reached on the study thus far. 20 patients of the 29 patients are still on treatment.
So, in conclusion, pembrolizumab has excellent activity in patients with heavily pre-treated Hodgkin lymphoma in a cohort of patients who all previously failed brentuximab vedotin. The CR rate was 21%, the overall response rate was 66%, the clinical benefit rate 25 of 29 patients. The safety profiles of these drugs are really quite good, very little grade 3 or 4 toxicity. Among the patients that we did stain for PD-L1 100% of the samples did stain for PD-L1.
I do believe that the results of this study should implore really both companies to continue our research with PD-1 inhibitors in a variety of subsets of patients with classical Hodgkin lymphoma.