Frontline Management of NSCLC After Approval of Durvalumab Plus Tremelimumab and Chemotherapy
Edward Garon, MD, David Geffen School of Medicine, University of California – Los Angeles, and Lyudmila Bazhenova, MD, University of California – San Diego, discuss options for frontline management of non-small cell lung cancer (NSCLC) after the most recent approval of durvalumab plus tremelimumab and chemotherapy based on results from the POSEIDON trial.
Transcript:
Edward Garon: Welcome to the Oncology Learning Network. My name is Edward Garon. I'm at the David Geffen School of Medicine at UCLA, and I'll be moderating today's discussion on options in frontline management for NSCLC. This is being recorded after the most recent approval in this approach, which is the combination of chemotherapy plus durvalumab and tremelimumab. This is based on the POSEIDON clinical trial. I'm thrilled to be joined by my colleague, Lyudmila, or Luda Bazhenova. Dr Bazhenova, maybe you can introduce yourself.
Lyudmila Bazhenova: Hello everybody. I am Dr Luda Bazhenova. I'm a thoracic medical oncologist and a lung cancer unit leader from University of California, San Diego.
Edward Garon: Thank you for joining us today. Of course, we have been in a situation where we've rapidly seen changes in the landscape for the frontline management of NSCLC, and this is now the third approved regimen of combination of a PD-1 or PD-L1 inhibitor, along with a CTLA-4 inhibitor. Maybe if you can go through the background on the data we have to date going into POSEIDON, related to the inclusion of CTLA-4 inhibition in the frontline management of NSCLC.
Lyudmila Bazhenova: Sure. The POSEIDON trial was a randomized phase 3 trial which compared 3 groups of patients: one is platinum doublet chemotherapy, another one is chemotherapy plus durvalumab, and the third arm was chemotherapy plus durvalumab plus tremelimumab. Statistical design of the study was that the durva-chemo arm and durva-treme-chemo arm were not designed to be statistically compared against one another. And it's basically a comparison of durva-chemo vs durva-treme-chemo vs chemotherapy.
I think this shows the difficulties of drug development in the current state where, by the time we finished the study, the standard of care moves on. And in the POSEIDON trial, chemotherapy was a comparison arm, but we know, ever since approval of immunotherapy as a monotherapy for PD-L1 >50, as well as PD-L1 >1, as well as approval of KEYNOTE-189, which is chemoimmunotherapy, chemo-pembro, as well as approval of ipi-nivo, based on CheckMate-227, as well as approval of ipi-nivo, chemo, based on CheckMate-9LA. We know that chemotherapy is no longer the standard of care, so this POSEIDON trial adds on to our pretty extensive body of knowledge that chemotherapy is no longer the right choice for our patients.
Edward Garon: Yeah. No, and I think that one of the things that you point out, of course, which is a complication here is that, although this is a comparative trial, both of the arms are compared to chemotherapy. There's no real formal comparison between the 2 arms. And as we get into the data, one of the things that we will clearly see is that one arm did hit its statistical significance threshold, that being the combination of chemotherapy, durvalumab, and tremelimumab, whereas the one with chemotherapy and durvalumab did not hit its end point. But again, the study was not really designed to formally compare those 2 arms, which is an interesting thing.
I'll just go through the design of the study. The study was a little over 1000 patients that were randomized to these 3 arms. As I mentioned, the control arm was platinum-based chemotherapy. It was histology-specific, meaning that this study did include both patients with squamous, as well as non-squamous NSCLC. You can see that the other arms were the combination of durvalumab and chemotherapy, as well as the combination of durvalumab, tremelimumab, and chemotherapy. The arm that both durvalumab and tremelimumab gave 4 cycles of the combination, and then there was one additional dose of tremelimumab that was given on week 16.
And for those who were receiving pemetrexed, those with non-squamous disease, of course this was an option, that group of patients was allowed to continue with the maintenance of pemetrexed and the durvalumab as well. Durvalumab was continued in both of the durvalumab-containing arms, but again, tremelimumab and maintenance was only given once at week 16.
And in terms of the results, statistically, the durvalumab, tremelimumab, and chemotherapy arm outperformed the platinum-based chemotherapy arm. Whereas, although there was a numeric superiority of the durvalumab plus chemotherapy arm, that did not have statistical significance in its benefit over the combination of platinum-based chemotherapy. This has now been approved in this setting.
This is a question maybe for you, Luda, now that we have really 3 different regimens that include a CTLA-4 inhibitor, how do you decide in whom you would be interested in giving CTLA-4 inhibition vs, for instance, immunotherapy alone with just single agent PD-1 inhibition, or combinations of PD-L1 inhibitors plus chemotherapy?
Lyudmila Bazhenova: This is a very good question and this showcases the challenges that we have when we practice oncology in real life is, commonly, we do not have a scientific answer to the questions that we are posing to ourselves. We do have a clinical trial ongoing called INSIGNA, which at least will answer the question of monotherapy immunotherapy vs monotherapy chemotherapy plus immunotherapy. And it'll also answer the question of continuation of immunotherapy if the patient progressed on monotherapy immunotherapy with an addition of chemo afterwards. But this INSIGNA trial will never answer the question of benefit of combinatorial immunotherapy with CTLA-4, as well as PD-L1, because the arm is not there.
In this situation, we have to do what's called a cross-trial comparison, which we are criticized for, but we have to make decisions somehow. And when you do the cross-trial comparisons, I think all those dual IO approvals, the numbers are very similar to the numbers of chemotherapy plus immunotherapy, and in some setting, very similar to monotherapy immunotherapy. However, what I think we are learning from the dual IO combinatorial trials is the potential benefit of CTLA-4 in alterations in STK11 and KEAP1. And I'm also a believer of a subset analysis of using CTLA-4 plus PD-L1, for patients with PD-L1 zero. Currently, I cannot use ipi-nivo for PD-L1 zero because it's not approved, so if I'm thinking of using dual IO therapy for PD-L1 zero patients, I am limited to CheckMate-9LA and POSEIDON.
And again, this is all a subset analysis. We do not have any randomized trials telling us that this is the right thing to do, but we have at least several subset analyses. We have a subset analysis from 9LA, from 227, from MYSTIC, as well as POSEIDON, telling us that patient with PD-L1 zero and patient with STK11, might benefit from addition of CTLA-4. And it does have a biological plausibility, because we know that patients with STK11 mutations do have what's called "cold tumors." So maybe addition of CTLA-4 does help.
And so, going back to the questions, what do I do? If I have a patient who have a PD-L1 >50, who does not have large disease burden or is not a never smoker or former light smoker, I tend to stick with immunotherapy by itself, if the patient is a never smoker, formal light smoker. And still, again, I make that decisions after I know what mutations they have, so it's not a decision upfront of the mutations. My never smoker patients in that population would've been tested for the molecular testing, and if they don't have anything, I still don't feel like immunotherapy by itself is a good idea for them, so large disease burden, never smoking situation, I tend to use chemo-IO, even for PD-L1 >50. PD-L1 zeros, I tend to use either CheckMate-9LA or POSEIDON. And then STK11, KEAP1s, also tend to use either CheckMate-9LA or POSEIDON.
Edward Garon: Well, thank you very much Dr Bazhenova for this excellent discussion. This concludes part 1, and I'd like to thank Oncology Learning Network for organizing the discussion. Please check out www.oncnet.com for part 2. There, we're going to be going more in depth into some of these biomarker-selected populations that Dr Bazhenova just raised in our discussion here. Thank you very much.
