Riess Reviews Pivotal Trials for Nonmetastatic Lung Cancer

Jonathan Riess, MD, MS

Associate Professor

Medical Director, Thoracic Oncology

UC Davis Comprehensive Cancer Center

Sacramento, CA

Targeted Oncology: Can you review the data for durvalumab (Imfinzi) in patients with locally advanced unresectable non–small cell lung cancer (NSCLC)?

RIESS: This is getting to the PACIFIC study [NCT02125461].¹ This was for [patients with] stage III locally advanced, unresectable disease treated with chemoradiation. [From] 1 to 42 days after chemoradiation, they were randomly assigned to durvalumab vs placebo for consolidation. In terms of the 5-year overall survival (OS) there was a clear difference: 43% [with durvalumab] vs 33% [with placebo], with a highly statistically significant HR, 0.72 [95% CI, 0.59-0.89].² With an end point of [OS] benefit, this was a landmark study. [Looking at] the 5-year update for progression-free survival [PFS], one-third of patients were without progression in 5 years with durvalumab vs 19% with placebo [HR, 0.55; 95% CI, 0.45-0.68].

[When looking at] the forest plot of OS by subgroup, I would highlight some of the differences.² I would highlight patients’ best response for prior treatment. [If patients] had partial response or stable disease, they both had benefit, interestingly [Table²]. [Patients who previously received] cisplatin seemed to do better than [those who received] carboplatin, whether that’s systemic doses or whether it’s just simply patient selection bias, or more fit patients.

It’s the same thing with [patients who were] randomly assigned to treatment with durvalumab less than 14 days since last radiation [benefiting more]. Aspirationally, that is the best, so try to start them within 2 weeks, but once again, there is patient selection bias. If you come out doing great after chemoradiation, you are probably thinking that you can start it right away, and you are probably going to do better… than somebody whose performance status becomes 2. That takes a while to recover from. Patients with EGFR [or ALK alteration] showed no clear benefit, in small numbers.² I would also highlight that those patients with PD-L1 expression of 0 [who did not benefit]; they have the durvalumab biomarker, which is different than pembrolizumab [Keytruda], but still very comparable. It is not approved in Europe for PD-L1 of 0, but it is here.

I give it because it’s hard to deny somebody with 40 pack-years of smoking the opportunity to benefit [because of] the subset analysis. Those are some of the nuances going through the subset analysis in the study.

The safety summary was comparable, a little bit higher for any-grade pneumonitis [33.9% with durvalumab vs 24.8% with placebo], but rates of grade 3 and 4 adverse events [AEs] were comparable.¹ There was a higher rate of any-grade pneumonia, as well [13.1% with durvalumab vs 7.7% with placebo]. But it’s a landmark study, a change in standard of care.

Can you discuss the real-world follow-up to this study?

Now we have the real-world data as well [PACIFIC-R; NCT03798535].³ Once again, [patients had] simple unresectable stage III disease, no evidence of progression following platinum-based chemoradiation. Data were extracted from patients’ medical records with retrospective data collection. It was looking at OS at 3 years, and then 5 years of the primary end point of PFS.

If you look at PFS and OS, the median real-world PFS was 21.7 months [95% CI, 19.1-24.5], 2-year real-world PFS was 48% [95% CI, 45.4%-50.9%] and 2-year OS was 71.2% [95% CI, 68.8%-73.6%].³ There were excellent outcomes in a real-world setting. The median time from end of radiotherapy to start of durvalumab was 56 days…. Approximately one-third started durvalumab within 42 days, with a median total treatment duration of 11 months.

Is there a correlation between time from end of radiotherapy to start of durvalumab and the outcome in the real-world data?

It’s a good point. The main issue is it’s going to be fraught with that patient selection bias, because [if you had] good performance status, you’re going to get started earlier. That’s the real challenge, in knowing what was true, if you start earlier. If you randomly assign patients to start earlier vs starting later, that would be a way to answer the question.

In terms of efficacy, in patients with PD-L1–negative disease, [PFS benefit was shorter], but it still seems to be a benefit.³ For disease stage, patients with stage IIIA did better than those with stage IIIB and IIIC. For tumor histologic type, nonsquamous did better than squamous histology; it would generally be better across the board. This real-world study included sequential chemoradiation, which required clinical trials. They had [shorter PFS than those with concurrent chemoradiation]. Once again, this is probably a patient selection issue as well. Patients who are less fit don’t even have the sequential approach. I don’t see that they did a comparison between start times [after radiation therapy]. But the median time [from end of radiation therapy] was 56 days, so everybody seems to be on the later side in the real-world setting.

Nearly half of patients [47.1%] completed treatment, with discontinuation due to disease progression at 27% and due to AEs at 17%.3 Then 47% had AEs [of special interest]. Less than 1 in 5 needed either a temporary or permanent discontinuation. Pneumonitis was a little bit higher [than in PACIFIC], 17.9%. It was the most common AE leading to interruption or permanent discontinuation, [which is] not surprising. That is the main toxicity we always worry about.

Can you review the data for osimertinib (Tagrisso) for patients with nonmetastatic NSCLC?

I think that we’re familiar with the ADAURA [NCT02511106] study, which took patients with [AJCC (American Joint Committee on Cancer) Cancer Staging Manual] 7th Edition, stage IB to IIIA NSCLC, surgically resected, with or without adjuvant chemotherapy, although that was highly encouraged in patients in whom it was indicated.⁴ Patients were randomly assigned to osimertinib vs placebo for EGFR exon 19 deletion or L858R. The treatment duration was 3 years with the primary end point of disease-free survival [DFS], and secondary end point of OS, among others. Following the IDMC [independent data monitoring committee] recommendation, given the DFS benefits, the study was unblinded. An interim analysis was reported that baseline characteristics were very similar in both groups, placebo vs osimertinib.⁴

Looking at their primary end point of DFS for patients with stage II to IIIA disease, the split in the Kaplan-Meier curves showed wide differences with an HR of 0.23 [95% CI, 0.18-0.30], highly statistically significant.⁵ When you throw in the stage IB disease, that also persists [HR, 0.27; 95% CI, 0.21-0.34] for a tremendous DFS benefit.

Looking at the subsets that persisted across stage, although stage III numerically had the best HR of 0.12 [95% CI, 0.07-0.20],⁴ you can see across all HRs, including all stages, including IB, the HR of 0.39 [95% CI, 0.18-0.76], still a tremendous benefit in DFS. [In] an updated presentation from ESMO [European Society for Medical Oncology Congress] last year, it showed that the benefits for the updated analysis persisted with 4 years’ DFS.⁵ Once again, there were numerically very impressive HRs for DFS benefit. This was a change in our standard of care. They then translated it to the 8th edition AJCC, and that, not surprisingly, persisted.⁵ Once again, stage IB had an impressive HR numerically, but it was really the stage IIIA disease that had the best numeric HR we could see. Across the board, it showed a tremendous DFS benefit. Part of the benefit is less CNS [central nervous system] disease [recurrence], so osimertinib had blood-brain barrier penetration.⁵ We know there’s less CNS disease in these patients. The recurrence HR was 0.24 [95% CI, 0.14-0.42].

There was a [news] release that OS [benefit was demonstrated and] would be presented at ASCO [American Society of Clinical Oncology Annual Meeting] this year.⁶ The trend was toward that good HR. [In patients with stage II-IIIA disease, median OS was not reached in either arm; HR was 0.49 favoring osimertinib; 95% CI, 0.33-0.73; P = .0004].⁷

The phase 3 ADAURA trial had the usual toxicities of rash and diarrhea, etc.⁵ Osimertinib can be used in the metastatic setting.

Can osimertinib potentially be a substitute for adjuvant chemotherapy?

I would say no. [Though there is] updated OS benefit with osimertinib, the study had adjuvant chemotherapy in eligible populations. Although it wasn’t mandated, it was highly encouraged. I do not consider it a substitute for adjuvant cisplatin-based chemotherapy, although there is a survival benefit. [If] you look at the curves [for] 3 years of osimertinib, when you stop it, you see that tick-tick down at 36 months of that DFS curve.⁵ There are patients recurring. I think there’s a different mechanism, more of a [cytostatic] mechanism.

But in the data…[there were] no half-CRs [complete responses]. It’s a different mechanism, you got that tick-tick—[with] multiple EGFR studies, going back to the first generation, when you stopped it and got that tick-tick-tick down DFS curve. Now, DFS benefit is so impressive, and you may be curing some of these patients, but it has a different mechanism of action. The long and short of it is, I don’t consider it a substitute for cisplatin-based chemotherapy. Now if a patient told me they refuse cisplatin-based chemotherapy, and they have an EGFR mutation, I would still [use osimertinib].

_REFERENCES

_{1. Antonia SJ, Villegas A, Daniel D, et al. Durvalumab after chemoradiotherapy in stage III non–small-cell lung cancer. N Engl J Med. 2017;377(20):1919-1929. doi:10.1056/NEJMoa1709937}

_{2. Spigel DR, Faivre-Finn C, Gray JE, et al. Five-year survival outcomes from the PACIFIC trial: durvalumab after chemoradiotherapy in stage III non–small-cell lung cancer. J Clin Oncol. 2022;40(12):1301-1311. Published correction appears in J Clin Oncol. 2022;40(17):1965.}

_{3. Girard N, Bar J, Garrido P, et al. Treatment characteristics and real-world progression-free survival in patients with unresectable stage III NSCLC who received durvalumab after chemoradiotherapy: findings from the PACIFIC-R study. J Thorac Oncol. 2023;18(2):181-193. doi:10.1016/j.jtho.2022.10.003}

_{4. Wu YL, Tsuboi M, He J, et al; ADUARA Investigators. Osimertinib in resected EGFR-mutated non–small-cell lung cancer. N Engl J Med. 2020;383(18):1711-1723. doi:10.1056/NEJMoa2027071}

_{5. Tsuboi M, Wu Y, GroheC, et al. LBA47 - osimertinib as adjuvant therapy in patients (pts) with resected EGFR-mutated (EGFRm) stage IB-IIIA non–small cell lung cancer (NSCLC): updated results from ADAURA. Ann Oncol. 2022;33(suppl 7):S808-S869. doi:10.1016/annonc/annonc1089}

_{6. Tagrisso demonstrated strong overall survival benefit in the ADAURA phase III trial for adjuvant treatment of patients with early-stage EGFR-mutated lung cancer. News release. AstraZeneca. March 9, 2023. Accessed August 23, 2023. https://tinyurl.com/5n7vk98f}

_{7. Herbst RS, Tsuboi M, John T, et al. Overall survival analysis from the ADAURA trial of adjuvant osimertinib in patients with resected EGFR-mutated (EGFRm) stage IB-IIIA non–small cell lung cancer (NSCLC). J Clin Oncol. 2023;41(suppl 17):LBA3. doi:10.1200/JCO.2023.41.17_suppl.LBA3}