I’m talking today about combinations of immunotherapies and that includes combinations of two different immunotherapies together, in fact two or more, but also combinations with more conventional treatments as well. I guess what I’ll be trying to do is to illustrate why there may be benefit in using combinations but also some of the pitfalls of looking at combinations and trying to convince your audience that when they’re thinking of designing their own clinical trials they need to think about things in a way that examines combinations that may result in synergistic benefit to a patient rather than simply additive benefit of the components of the combinations which is a bit of an issue at the moment.

Can you give some examples of these combinations?

So combinations of immunotherapies have now become standard of care in the two diseases that I treat, melanoma and kidney cancer. Combination ipilimumab and nivolumab is standard of treatment for advanced melanoma and I’ll be presenting data, and James Larkin presented a bit of it today, with the BMS 214 study showing combination ipi and nivo in kidney cancer is also significantly better than conventional treatment with a tyrosine kinase inhibitor. So in those two diseases combinations of two immunotherapy drugs, two checkpoint inhibitors, have become standard of care because they’ve shown themselves to be superior to their comparators.

Do you get problems with toxicity when combining these two therapies?

You can do, that’s an important element to the question. What we’re learning is that the toxicity profile is very much dose and schedule dependent of the drugs that we’re looking at. The dose of ipilimumab for melanoma in combination is higher than the dose of ipilimumab in the renal cancer schedule and we see differing toxicities as a result of that. One of the issues about toxicity is what the purpose of treatment is. So patients who are treated with high dose chemotherapy with metastatic germ cell tumours have very severe toxicities but they’re treated with curative intent. You justify the toxicities because you’re treating with curative intent. Actually that’s now the case with immunotherapy for melanoma and kidney cancer, it may not be the case for immunotherapy yet with some of the other more common diseases. So the efficacy signal is an important component of considering whether a toxicity is justifiable or not.

You also mentioned other pitfalls.

The other issue is really about trial design and readout because it comes back to this key point about why we’re giving immunotherapy in the first place and we’re not giving immunotherapy to try and gain short-term palliation of patients like my colleagues who treat epithelial cancers with conventional chemotherapy do. What we’re trying to do is to get long-term durable disease control and actually I’m after my patients not dying from their cancer, that’s my goal. So that means that the goal is long-term benefit but the trial readouts are very often short term and it may well be that short-term readouts are not representative of the long-term benefit. So you can make premature and flawed decisions about the value of drugs because you’re not looking late enough and you’re looking at endpoints that are too early. Now that has to be balanced against obvious commercial and clinical pressures to get active drugs licensed and reimbursed as soon as possible so there is a tension there between the pressure to get things out early but where the efficacy signal, the important element of the efficacy signal, might not be most apparent until later.

Do you ever have patients requesting treatment that they’ve heard about?

Yes we do a lot and actually I embrace those conversations and the more somebody comes in having done some legwork on what’s going on the higher quality, generally, of the consultation. The reality is that we are inevitably restricted by NICE approval and reimbursement decisions in the UK. I’m not saying that’s a bad thing, it’s just a reality check, so there are often limits on what we can do. But I also think that patients have a right to be aware of where the data is heading and also what I think is the best treatment for them. I will describe that in a way that takes into account of what’s available on the NHS but I will also be explicit about what may be coming or what isn’t available on the NHS because I feel when people come to me for advice they’re coming for my advice about the field and not advice about NHS reimbursement decisions.

Key message for a doctor watching this?

Key message would be that it’s a fascinating, exciting field. Combinations will be the standard of care but there are so many drugs there that the way we assess combination immunotherapies needs to be quite thoughtful and we need something I haven’t mentioned so far which is high end translational work to really instruct which patients are going to have benefit from which combinations. At the moment we have patients who have long-term benefit and many who don’t, the question is can we be a bit more selective about who we treat with the currently available drugs and who may require some of the newer combinations. That’s essential.