Mark A. Socinski, MD: We saw the recent results of overall survival [OS] in the PACIFIC trial. Suresh, I’m going to ask you to walk us through this and provide your thoughts on the new standard of care. From a radiation oncology point of view, how to you see this data?

Suresh Senan, MRCP, FRCR, PhD: It’s an exciting development for the standard of chemoradiotherapy and patients who are not progressing and receiving durvalumab [Imfinzi] or placebo 2 in 1 fashion. It was encouraging to see the OS readout with a hazard ratio of 0.68.

Mark A. Socinski, MD: Pretty impressive.

Suresh Senan, MRCP, FRCR, PhD: Very impressive. The progression-free survival [PFS] has been maintained. There were some concerns about the trial, but it’s reassuring that there are no safety issues, especially for the lungs, as we do the follow-up. Interestingly, it’s also important to note that the standard arm’s median overall survival came within the range of what you would expect.

Mark A. Socinski, MD: 28 months, right?

Suresh Senan, MRCP, FRCR, PhD: 28.7 months or something like that. Durvalumab is not looking good because of underperformance in the standard arm.

Mark A. Socinski, MD: If anything, it was better than we would have anticipated.

Suresh Senan, MRCP, FRCR, PhD: The presentations at World Congress didn’t shed much light on why the hazard ratio was more favorable when we give it within 14 days. Is it because the small tumors cause less collateral damage and recovery time, or is it because of some other biological factors?

Mark A. Socinski, MD: There’s some magic to giving it early.

Suresh Senan, MRCP, FRCR, PhD: It seems so. If you look at the animal data, it’s more advantageous to give it as soon as possible. That’s something that in clinical practice, we’re trying to see whether we can shorten that gap, in this context, with a PD-L1 [programmed death-ligand 1] inhibitor.

Mark A. Socinski, MD: One of the concerns with PACIFIC was the interaction to immune-related pneumonitis, realizing that radiation pneumonitis is an issue. The data from PACIFIC were reassuring that we aren’t exposing patients to undue risk in that setting.

Suresh Senan, MRCP, FRCR, PhD: I would agree with that. Grade 1 and 2 pneumonitis was a little higher, especially in the East Asian populations—but in the rest it was very reassuring. And high grade, grade 3, 4, and 5, were very manageable.

Mark A. Socinski, MD: I think 3%, but it was 3% on placebo also.

Suresh Senan, MRCP, FRCR, PhD: Exactly.

Mark A. Socinski, MD: We don’t know the dose volume histogram data with regard to dosing of the lung.

Suresh Senan, MRCP, FRCR, PhD: Some of the data have come out, but it’s an international trial with many patients; it is also incomplete. What did it treat? That is what the community is looking for: who has not been treated and who did come through to the randomization. It means we need better-designed trials to tease out some of these factors. Of course, the biomarker issue is also there.

Maximilian Hochmair, MD: What I like about the PACIFIC trial is that it’s real-world data because it was not a special chemotherapy or radiotherapy technique. It showed these great results. What is new with this trial also, before we tested PD-L1 staining for patients in stage IV, is that we now have to put it in earlier stages. We are testing PD-L1 for those patients because we saw good results for the PD-L1 staining patients.

Mark A. Socinski, MD: Something that’s interesting to me is you know that the sooner you give durvalumab after a completion of chemoradiation, the better off the patient may be. In the United States, there has been reluctance to give up consolidation chemotherapy, which is surprising to me because we were never convinced that either induction or consolidation made a difference, but somehow people subscribed to the idea that you have to give consolidation. I like induction, and the PACIFIC trial has, in my opinion, given me permission to administer induction therapy since 28% of the patients in the PACIFIC trial did receive induction. Stage III is still a 4-cycle disease. You know we have to give an adequate systemic exposure of chemotherapy because we’re dealing with local regional disease as well as micrometastatic disease. I will admit, however, that I cannot prove this.

Maximilian Hochmair, MD: I’m completely in line with you.

Mark A. Socinski, MD: I knew you were a smart guy.

Maximilian Hochmair, MD: For me it was strange that you start an I-O immediately after radiotherapy. The patient is coming from durvalumab in this situation. This in a completely practice-changing study.

Mark A. Socinski, MD: Historically we’ve had a lot of therapeutic nihilism associated with lung cancer. To me, seeing this data provide a heightened level of enthusiasm to make sure those patients get treated.

Thomas E. Stinchcombe, MD: It does improve the stage III and the popularity of this therapy when you start seeing prolonged survivals. The time to distant metastases was positive as well—it’s really the cause of systemic therapy.

Mark A. Socinski, MD: I did not need the OS curve that we saw at the World Conference on Lung Cancer [WCLC]. I saw either time to death or time after first metastasis, which to me is a poor man’s Kaplan-Meier OS. What we’re trying to do with chemoradiotherapy is prevent stage IV disease and prevent death. The hazard ratio for that was initially close to 0.5. To me that was enough to say there’s a survival advantage.

Maximilian Hochmair, MD: And it’s a systemic effect.

Leora Horn, MD, MSc, FRCPC: In the United States the restriction is not based on PD-L1 status.

Mark A. Socinski, MD: That’s a good point.

Leora Horn, MD, MSc, FRCPC: We don’t test our patients often in early stages—the argument being, if they recur, you’re going to biopsy them, and you’re going to find out at that time. With the metastatic data, PD-L1 has never been great in predicting responders versus nonresponders. In the United States, you should use the data because the subset analysis was positive. Sometimes you don’t use it when the subset analysis is negative. It’s not restricted at this point to who you can give the drug to.

Mark A. Socinski, MD: In Europe it is.

Maximilian Hochmair, MD: I’m completely in line with you because in only 60% of the patients could we get PD-L1 staining; it’s not very practical. Even if it’s a real-world starting with highly educated centers, they got in 6% to 65% of the patients for PD-L1 staining. Your way is much better. It’s the United States label, so we have to do it.

Thomas E. Stinchcombe, MD: I have some hesitancies. We check for PD-L1. I don’t know how to apply it because we do it in everyone. I have a hesitancy of foregoing this therapy for a dubious biomarker and a small subset in which it may be that prognostic factors weren’t balances. We are treating these patients with curative intent; the threshold to abandon therapy needs to be high.

Leora Horn, MD, MSc, FRCPC: I abandoned induction. We were doing cisplatin-etoposide and weekly carboplatin [Paraplatin] with Taxol [paclitaxel]. Those patients couldn’t tolerate cisplatin-etoposide, so I had to abandon induction.

Mark A. Socinski, MD: If you’re using full-dose CIS-ETOP, that’s understandable. I have a hard time with patients who have a good performance status, 0 or 1, whose goal is cure by giving 6 weeks of low-dose carboplatin and Taxol as their only systemic exposure. Maybe you can get away with that if you use I-O afterward. We actually like induction; it takes the pressure off getting the radiation plan done. We have 6 weeks to do it. There’s almost always some degree of volume-metric reduction. My colleagues here published a paper years ago suggesting that preradiotherapy volume is a predictor of long-term survival. Volume is important with regard to radiation therapy.

Leora Horn, MD, MSc, FRCPC: Although Everett’s study of the induction of carboplatin and Taxol was positive.

Mark A. Socinski, MD: I understand that. I’m not sure you lived through Everett’s study, but I did.

Leora Horn, MD, MSc, FRCPC: I read it though.

Mark A. Socinski, MD: I don’t take CALGB 39801 as the true test of induction therapy; it was woefully underpowered. There was a 2-month difference in median OS that did not achieve significant difference, but the statistical design was looking for a 40% increase in OS. It was a big hurdle to overcome. Thank you for bringing that trial up.

Thomas E. Stinchcombe, MD: We also looked back at that trial: the median OS in both arms was terrible.

Mark A. Socinski, MD: There were many people in CALGB at the time who would not put a patient with good performance status—a good candidate for chemotherapy—on that trial because they didn’t want to have the patient get 6 weeks of low-dose carboplatin and Taxol. I was one of those people.

Transcript Edited for Clarity