Practice Pearls: Checkpoint Inhibitors for Lung Cancer

Mark G. Kris, MD


February 23, 2017

This feature requires the newest version of Flash. You can download it here.

Mark Kris here, talking about changes in treatment from 2016 for patients with lung cancers.

I think the single biggest change over the past year has been the emergence of immunotherapies, particularly agents targeting PD-1 and PD-L1. We have three FDA-approved ones now. In the second-line space, we have one already approved and others heading that way, and then the upfront space. These are drugs that have a place in the treatment for just about everyone, and not only do they have a place in adenocarcinomas and squamous carcinomas, they have a place in small cell lung cancer as well.

Selecting Patients

What about these drugs? Can you select which patients are most likely to benefit from them? It is clear with PD-L1 that for the initial therapy, the 50% cutoff of PD-L1 expression is our current standard, and that is the test that we perform today to determine whether you give a PD-L1 drug up-front versus administering it second line after chemotherapy.

The other test that is going to be increasingly important is mutational burden. Many of the next-generation platforms now give you mutational burden score, and we know from the work of [Dr Naiyer] Rizvi and [Dr Matthew] Hellmann and others that having a high mutational score means a greater chance of benefit from these agents. It is not perfect. It is not a guarantee of benefit. It is nowhere near as potent a predictive factor for benefit than, say, lack of EGFR mutation, but it is a good one. Very low expression, very low benefit; higher expression, higher benefit, both in terms of PD-L1 and of mutational burden.

Putting those two things together for individual patients is probably the best thing you can do. There are some surrogates for mutational burden; one is smoking status. More smoking, and longer smoking, is more likely to have high mutational burden. Those of you who look at the next-generation [sequencing] reports of patients with more smoking history will see that they have more mutations. It is just a measure of that.

Another surrogate is KRAS mutations. Many of them are associated with smoking. These are not perfect, but they are all hints as to the benefit from PD-L1 therapies.

Choosing a PD-L1 Agent

In prioritizing PD-L1 therapies where multiple choices are available, these factors are important. A critical issue with PD-L1 therapy is being able to stop it as soon as it is not helping. These drugs have side effects. Compared with chemotherapy, their side effects are better, but there are side effects nonetheless. Much has been said about the concept of pseudoprogression. It does happen, it is extremely rare, and it is definitely a single-digit incidence. The critical thing in order to tell pseudoprogression from what I call pseudo-pseudoprogression, or real progression, is the presence of symptoms. When you have a patient who has symptomatic progression, coupled with radiographic progression, on a PD-L1 agent—particularly a patient who has toxicity and a patient with low PD-L1 expression, low mutational burden, low smoking burden, and no KRAS mutation present—almost certainly that patient is suffering disease progression and should be offered a therapy with a greater likelihood of benefit.

These things are not absolutes. There is no absolute test to determine progression or to determine the likelihood of benefit to a PD-L1 drug. It requires a lot of thought. It requires amassing information, and I urge you to try to do your best with every individual patient to decide whether a drug is helping them.

Managing Side Effects

A couple of reminders about the unusual side effects. I must say that I never paid a lot of attention to thyroid disease. I pay a lot of attention to it now. For patients on PD-L1 drugs, a consensus recommendation would be that about once a month you would check the thyroid-stimulating hormone (TSH) level. When you follow these TSH levels, very often you see, early in the course, that TSH goes down. A hyperthyroid, at least TSH-based hyperthyroid state, is rarely clinically apparent; I have not seen a case of that. But then, [potentially] hypothyroid following that. I must say that I have never had a very clear-cut case of symptomatic hypothyroidism, but I have been told that they are out there, and a rule of thumb is: Once the TSH starts rising and is over 10, there should be some treatment with Synthroid.

Particularly for those patients getting an anti-CTLA-4 inhibitor with their PD-L1 drug, beware of pituitary insufficiency. It can be a little insidious. You know that our people with lung cancers are always complaining of fatigue, but when their fatigue becomes more bothersome and the patient says, "Doc, things are different now," please make sure you are considering pituitary failure. Document it and start supplementing that patient with hydrocortisone. Generally for central failure, treatment with a cortisone is good enough. You do not need to add mineralocorticoid, but again you must carefully look at those patients.

Diabetes can happen as well—a different kind of diabetes, and it is an autoimmune diabetes. People do not make insulin and they need insulin. Things like metformin are probably much less helpful here. This is a case where you could use the help of an endocrinologist. I would ask you to do that. People can also get myositis and arthritis. Be very careful to look for that. You can generally document the myositis by looking at a creatine phosphokinase blood test, looking at an aldolase test. Those patients require holding the drug depending on the severity of symptoms. Also, these side effects of the -itises are steroid-responsive, and you can always stop the drug and give steroids to get the patient through those crises.

It is a brave new world with immunotherapy. It's hard to know who exactly it is going to work in. It takes some work to figure it out. It's hard to know whether people have progressed or not. It takes a little work to figure that out—not so much about more side effects, but a different spectrum of side effects. These are not the usual things we meet about. Getting that CBC on everybody who walks in the door is not going to cut it for people on PD-L1 and anti-CTLA-4 drugs. It is a different set of tests you need. It is a re-education for ourselves, a re-education for our staff, and also for the patient who may have been used to getting a CBC all the time when on chemotherapy. Actually, the patient may wonder why you are not doing CBCs anymore.

It's good that we have these drugs. A lot of patients are helped. It gives us another option for just about everybody. We do have a harder job ahead, but a better one.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: