WCLC 2021 | Extending durvalumab maintenance to special populations

Immune checkpoint blockade has certainly revolutionized the treatment of advanced non-small cell lung cancer. But now we are moving these therapies also in the locally advanced setting and durvalumab is approved as the maintenance treatment post-concurrent chemoradiotherapy for patients who did not progress after two cycles of platinum-based concurrent chemorad. Recent data that were presented at ESMO with four years follow-up, show that there is a longer overall survival for patients receiving durvalumab over placebo, with a median survival of nearly 48 months. What is very interesting is that the patients who received the durvalumab have four years overall survival rate of nearly 50% compared to 36%, so this is a very active drug.

As we are rolling out these therapies in the standard clinical practice, so moving away from the registration studies, we are facing a number of clinical scenarios, which are involving populations which were not enrolled in the registration study. In particular, I’m referring to the elderly populations or the patients with pre-existing autoimmune disease, or patients with chronic viral infections such as hepatitis B or C or HIV.

There is a clear benefit from durvalumab maintenance, but what do we do if, for example, we have a patient who is an elderly patient? This is very relevant for lung cancer because vast majority of patients with lung cancer will be 65 plus. Well, the first thing is that there is no standardized definition of ‘elderly population’ in clinical trials. That’s because different trials have used different cutoffs, for example, 65 years old or 70. To my opinion, and this is shared by the oncology community, chronological age itself is much less important than biological events that are driving the aging process. As a most commonly used cutoff, for example, from the International Society for Geriatric Oncology, we are using 70 years old. But the vast majority of the trials which have reported subgroup analysis for elderly patients so far are still using 65, which you may argue is still a pretty young population.

One of the issues in using immune checkpoint blockade, or immunotherapy in general, for elderly patient is that aging is associated with the altered or decreased immune system functions across all the components of the immunity cycle. One of the concerns was that this drug were going to be less effective in the elderly population, and there are potentially different side effects, or more adverse events in these patients. Most of this data on the subgroup of patients who are elderly comes from the advanced studies or some subgroup analysis from not randomized studies, and therefore we need to extrapolate for durvalumab in the maintenance setting. What we know is that in the pre-treated population, patients who are elderly will derive the same benefit in overall survival when compared to the patients who are young. This is seen also in most recent first-line studies together with not randomized studies, such as CheckMate 171 where patients with pre-treated squamous cell carcinoma received nivolumab until progressive disease, and the intention-to-treat population had the same overall survival as the subgroup of patients above 70 years of age, which was more than 270 patients, so a pretty big sample.

This is something that we have seen also in some expanded access programs, which have been published in the literature. Very relevant is some data from the immunology group at Gustave Roussy who have been able to develop a senescent immunotype signature in blood with a specific phenotype of T-lymphocytes. In their study, they were able to observe 37 patients before having immunotherapy and out of these about 20% were considered to be senescent immune phenotype-positive, but very interestingly 58% of the senescent immune phenotype positive patients were actually young, so less than 65 years old. When they presented the data on disease control rate and progression-free survival for these patients after immunotherapy, you could clearly see that patients with senescent immunophenotype did less well in terms of disease control rate and progression-free survival.

Why is this data relevant? Because I believe if it’s validated prospectively, this could give us another tool to select patients who are most likely to benefit from immunotherapy independently on age. Now regarding the issue of autoimmune disease and lung cancer patients. We know that this is pretty common in patients with lung cancer. From the literature, up to a quarter of patients with lung cancer will have autoimmune disease, they’re most likely to be female, they have usually early stage disease and greater comorbidity burden, and the most common ones are rheumatoid arthritis, psoriasis and polymyalgia rheumatica. These patients were excluded from the registration studies with immunotherapy, including PACIFIC, which is the registration study for durvalumab.

The definition in these protocols was that patients with active or previous autoimmune disease, which was active within the past two years, couldn’t go on a study. The data that we have on these patients is mostly from retrospective series. What we know from a handful of series including non-small cell lung cancer patients, is that the risk of flare of autoimmune disease is low, and these patients seems to derive the same benefits in terms of response rate and progression-free survival than patients without the autoimmune disease. Therefore, I feel that immune checkpoint blockade such as durvalumab, for example, could be offered to patients with no life-threatening autoimmune disease or inactive, but you need to monitor your patients quite strictly.

There is a systematic review which involves patients with melanoma and lung cancer treated with ipilimumab and then anti PD-1 antibody. In this series, about 50% of the patients who were treated with immunotherapy had a flare, but the discontinuation rates related to these flare was only in the regional 17/18%.

Now, very important is the presence of a chronic viral infection, such as chronic hepatitis B or C or HIV. These patients were also excluded from the registration studies with immunotherapy. The exclusion of the patients with hepatitis B and C was mostly due to the fact that there was a concern due to the potential risk of hyper-immune response in these patients causing hepatic failure. Then it was difficult to tease out, for example, if the liver function test derangement was due to a reactivation of the hepatitis or immune-related hepatitis. Because these patients were not enrolled in the registration study, the data is pretty scant and is mostly from either retrospective series or studies involving hepatocellular carcinoma, where we have a lot of patients who are HBV or HCV positivity.

In these studies, for example, durvalumab was trialed in there, but hepatocellular carcinoma, what was important to see was that in patients with hepatitis, the disease control rate and the response rate was similar between the infected and non-infected group, and there was no difference in toxicity. There are some small series in non-small cell lung cancer. Again, mirroring what we have seen in hepatocellular carcinoma, where the response rate were very similar and also the immune-related adverse events rate were very similar in the HBV-positive or -negative study patients. From that point of view, I think is reassuring to see that if you have a patient with well controlled chronic hepatitis B or C, you may treat this patient as long as you manage them with the infectious disease team, and you keep a close eye on the adverse events.

HIV population is very relevant for lung cancer clinicians. This is because we know that patients with HIV will have an increased risk of developing lung cancer during their lifetime. In particular, if they are a smoker. The data again is pretty scant because these patients couldn’t go into any registration study, and there is a couple of series and a systematic review, which show that patients with different disease groups who were treated with immunotherapy had not a great deal of side effects. The severe immune-related adverse events was in the region of 8% in terms of rate and, very importantly for clinical practice, the HIV remained suppressed in more than 90% of the patients, and the HIV did not affect the efficacy of the immunotherapy, which is what you want to know when you’re embarking on these therapies in clinical practice.

There’s a couple of studies with durvalumab and the one with the… Sorry, one study with durvalumab and one with pembrolizumab, in patients with different kinds of cancers and HIV. They both showed that the toxicity was very minimal and, very importantly, there was no viral reactivation observed in this study. There’s also some preclinical data that shows that PD-L1 blockade could be very relevant for patients with HIV from an infectious disease point of view, as it may enhance the HIV-specific immunity, and it may help to purge the HIV reservoirs that will indefinitely persist, even in patients who are on antiretroviral. We do advise to continue monitoring the CD4 and CD8 count, and strictly manage these patients with the HIV clinicians if they’re on the immunotherapy.

In conclusion, I think, given the benefit of durvalumab, which is now a standard of care post concurrent chemo rad for inoperable locally advanced non-small cell lung cancer patients, we should consider patients who are elderly but fit and no contraindication to immunotherapy. I feel, also, that patients with inactive autoimmune disease and controlled chronic viral infection can be considered for PD-L1 blockade, as the potential benefits will definitely outweigh the risks related to toxicity with these patients.