Immunotherapy and Chemotherapy in Second-Line NSCLC
Benjamin P. Levy: If we look across the board at immunotherapeutic strategies—either nivolumab, pembrolizumab, or atezolizumab—they’ve all been compared to docetaxel second-line. Across the board, in the intention-to-treat analysis for these trials, immunotherapy elicits higher response rates than single-agent docetaxel. Again, in the intention-to-treat analysis, there’s no doubt that single-agent checkpoint inhibitors improve response rates, improve progression-free survival, and improve overall survival in all of these trials. Things get a little more interesting when you break down some of the subsets from these trials to look at who are the ones who actually responded or not.
For instance, in the CheckMate-057 study that compared nivolumab to docetaxel, if you look at the patients who had low PD-L1, there was no difference in survival between Opdivo [nivolumab] and docetaxel. Opdivo, nivolumab, is obviously better tolerated than docetaxel. This is a welcome change. This has altered the treatment paradigm for patients. Immunotherapy is a standard of care for patients who are platinum refractory, patients who have progressed on platinum chemotherapy. There are nuances in these trials that we have to look at. I also say that the higher the PD-L1 was for many of these trials—outside of the squamous cell trial, which was CheckMate-017—the higher the PD-L1 was for these patients, the more likely they were to respond to immunotherapy. So, I think that’s important to note as we begin to think about second-line decisions for our patients.
This is where things get a little more interesting. For patients who have rapidly progressing disease, if you look at the subset analysis from CheckMate-057, this was the trial comparing nivolumab to docetaxel in lung adenocarcinoma. In that trial, there was no doubt that nivolumab improved survival over docetaxel. But if you look at the subset analysis for patients who actually had rapidly progressing disease, for patients who had poor prognostic features, and for patients who had low PD-L1 expression, there were actually more deaths in the Opdivo arm, the nivolumab arm, than there were in the docetaxel arm in the first 3 months.
This is a scenario that I think has been underrepresented in the literature in terms of what we do. I think this is a scenario of a patient who’s rapidly progressing. While we generally just reflexively use immunotherapy for all patients second-line, for patients who are rapidly progressing, who are symptomatic, and who have a low PD-L1 score, these are the patients where maybe immunotherapy is not the best option. And there are certain scenarios we can talk about where immunotherapy is not the best option. This may be one of them, and we’ll talk about some other scenarios where that may be the case. So, there should be special consideration for these patients for second-line options.
This is the other scenario that we really have to be careful of with immunotherapy. Again, the TV commercials and the advertising campaigns would make you think that immunotherapy works for all lung cancer patients. But if we look specifically at patients with EGFR-mutant lung cancer, if we look specifically at the ALK rearranged lung cancer patients, across the board, immunotherapy has been less efficacious than docetaxel. And that’s important. I don’t think that message has really gotten out there. In the subset analysis of the OAK trial, patients who had EGFR mutations did better with docetaxel than they did with immunotherapy.
In the subset analysis of the CheckMate-057 trial, it was the same thing. Patients with EGFR-mutant lung cancer responded better to docetaxel than they did with immunotherapy. This is another situation and another clinical scenario where, if they have an EGFR mutation or if they have an ALK rearrangement and these patients get targeted therapies and then they get chemotherapy, I think there’s a reflex to say, “OK, the next step is immunotherapy.” And I would argue maybe not.
There have been some retrospective analyses to suggest the response rate to immunotherapy for patients with EGFR-mutant lung cancer is only around 5% to 10%. The response rate for ALK rearranged lung cancers to immunotherapy is 0%—very small numbers. We’ll find out soon in well-designed trials about the role of immunotherapy. I’m not saying immunotherapy is out for these patients, I’m just saying that we need more data. There are some good trials that are being done looking at the role of immunotherapy in these genotypes that I think will help better answer this question.
If I’m going to use immunotherapy up front with chemotherapy in a treatment-naive patient, we really don’t have many options because we’ve already used immunotherapy up front. I’m not saying that this is not a good strategy to use—triplet therapy with carboplatin, pemetrexed, and pembrolizumab. But let’s say for these patients that they do progress: I think, for a nondriver mutation–positive patient, there are very few options left.
One good option would be docetaxel with or without ramucirumab. For a patient who does progress on triplet therapy with carboplatin, pemetrexed, and pembrolizumab, second-line, if that patient is fit, I would use docetaxel and ramucirumab. If the patient was not fit or had lots of toxicities, I would maybe use single-agent gemcitabine or maybe weekly docetaxel without the ramucirumab. So, again, treatment decisions in second-line have to be individualized, and you really have to look at the side effects of a drug, look at it in the context of the advantages it provides in terms of response rate and overall survival.