CME
Physicians: Maximum of 1.25 AMA PRA Category 1 Credits™
Released: September 01, 2022
Expiration: August 31, 2023
Stephen V. Liu, MD:
Although we have seen clear improvements in treatment options for advanced and metastatic NSCLC, there remains room for improvement. There were a couple of important updates on disease management in this setting at ASCO 2022.
Stephen V. Liu, MD:
Brain metastases in NSCLC are associated with poor prognosis and quality of life.12 Local treatment of brain metastases can add toxicity and delay systemic treatment for the primary cancer.13 Immune checkpoint inhibitors (ICI) alone or with chemotherapy show promising intracranial efficacy and safety.14,15 However, patients with untreated brain metastases and those who receive corticosteroids have been underrepresented in clinical trials evaluating first-line IO in combination with chemotherapy.16
The ATEZO‑BRAIN study is a nonrandomized, single‑arm phase II trial for treatment-naive patients with nonsquamous stage IV lung cancer and untreated brain metastases.17 Patients were eligible if they had either no neurologic symptoms or symptoms controlled with anticonvulsants or low-dose dexamethasone and were EGFR/ALK negative with any PD-L1 expression level. All patients received carboplatin, pemetrexed, and atezolizumab every 3 weeks for 4-6 cycles. The coprimary endpoints were safety, investigator-based PFS by Response Evaluation Criteria in Solid Tumors (RECIST) v1.1, and response assessment in neuro-oncology brain metastases (RANO-BM). Secondary endpoints included ORR, OS, quality of life, time to brain radiotherapy, and neurocognitive function.
Stephen V. Liu, MD:
Eighty-five percent of enrolled patients were current or former smokers, and at baseline, more than one half (55%) of patients were receiving corticosteroid therapy.17 One half of patients in this trial had tumors with PD-L1 expression between ≥1%.
Stephen V. Liu, MD:
Grade 3/4 TRAEs occurred in 70% of patients and serious TRAEs in 7 (17.5%) patients: acute kidney injury (n = 2), nephritis (n = 1), pneumonitis (n = 2), febrile neutropenia (n = 1), pulmonary embolism (n = 1).17 One patient (2.5%) had a grade 5 TRAE (febrile neutropenia and infection).
Neurologic AEs reported in the ATEZO-BRAIN trial included headache, dizziness, seizure, and insomnia. One patient (2.5%) experienced grade 4 hallucinations, 2 patients (5%) had grade 3 seizures, and 2 (5%) had other grade 3 nervous system disorders including sciatica and spinal cord compression. There were more neurologic AEs observed in ATEZO-BRAIN than other studies, such as CheckMate 9LA and KEYNOTE-021/-189/-407, have noted, but these were not all related to treatment and a lot of these were related to the underlying brain metastases.15,18
Stephen V. Liu, MD:
The intracranial ORR using RANO‑BM criteria was 42.5%, and the systemic ORR by RECIST criteria was 45%, suggesting high concordance between brain and body.17 This is higher than previous studies of ICIs alone. Although we do not know which therapies are responsible, or whether it is all 3, we are clearly seeing some favorable responses within the brain with this regimen.
Stephen V. Liu, MD:
With this update to the ATEZO-BRAIN trial, the median systemic PFS by RECIST was 8.9 months (95% CI: 6.7-13.8) and intracranial PFS by RANO-BM was 6.9 months (95% CI: 4.7-11.9).17 The median OS was 11.8 months (95% CI: 7.6-16.9) and the 2-year OS was 27.5%.
Stephen V. Liu, MD:
The median time to brain radiotherapy was 10.9 months (95% CI: 7.8-15.9). In this trial, EFS was defined as the time to brain radiotherapy or death, and the median EFS was 7.62 months (95% CI: 5.5-10.9). The 3‑month EFS rate was 87.2%, although the EFS rate fell to 25.6% at 12 months.17 Although most (60%) patients did move on to radiation therapy, this was not immediate, and clearly this regimen is an effective strategy for some patients.
Stephen V. Liu, MD:
Brain metastases are common in lung cancer, and I think these data provide support for using systemic chemotherapy/IO in driver mutation–negative lung cancer to control central nervous system (CNS) disease. I think the key with this strategy is patient selection.
Zofia Piotrowska, MD, MHS:
I agree. These data confirm what we have seen in clinic. We know that carboplatin and pemetrexed have some CNS activity, and I think we anticipated that adding IO to that combination would likely be at least as good if not a little bit better in terms of CNS activity. These response rates are modest, however, and I think it is important to remember this approach is not the same as having a CNS‑penetrant TKI. This atezolizumab plus carboplatin/pemetrexed regimen may be an appropriate strategy for patients with small, asymptomatic brain metastases, but for patients with more symptomatic or larger brain metastases, I think that reviewing them in a multidisciplinary fashion with radiation oncology and adding in radiation early remains critical.
Stephen V. Liu, MD:
I agree this is not appropriate for everyone with untreated brain metastases but is an option for some.
Stephen V. Liu, MD:
Eftilagimod alpha is a recombinant, soluble LAG-3 fusion protein that binds to major histocompatibility complex class II molecules to mediate the activation of antigen-presenting cells.19,20 Activation of antigen-presenting cells leads to CD8 T-cell activation, which potentially reduces the number of nonresponders to PD-1/PD-L1 antagonists.
TACTI-002 is a multicohort phase II trial of eftilagimod alpha plus pembrolizumab in patients with untreated advanced/metastatic NSCLC, recurrent/refractory NSCLC, or recurrent/metastatic squamous head and neck cancer.21 The data that were presented at ASCO 2022 were from the cohort of patients with previously untreated NSCLC, not amenable to ALK/EGFR inhibitor therapy or treatment with curative intent (Part A). Patients received a combination of pembrolizumab every 3 weeks with eftilagimod alpha every 2 weeks for the first 8 cycles, then both treatments every 3 weeks for 9 cycles up to 1 year, followed by 16 cycles of pembrolizumab monotherapy for up to 1 year. The primary endpoint was ORR per Immune Response Evaluation Criteria in Solid Tumors (iRECIST). Key secondary endpoints included ORR by RECIST v1.1, PFS, OS, duration of response (DoR), and safety.
Stephen V. Liu, MD:
In this trial, 95% of patients were current or former smokers and most (63%) had nonsquamous tumor histology. Patients were previously treated with radiotherapy (33%), surgery (20%), or systemic therapy for nonmetastatic disease (22%). The study enrolled patients across all PD‑L1 strata, generally reflecting what we see in the clinic. Approximately one third (33%) of patients had PD‑L1–negative tumors (TPS <1%), 35% of patients had PD‑L1–low tumors (TPS 1%-49%), and 27% had PD‑L1–high tumors (TPS ≥50%).21
Stephen V. Liu, MD:
Overall, response rates with this chemotherapy‑free regimen were modest. ORR was 38.6% by iRECIST and 37.7% by RECIST v1.1.21 Most responses were partial responses (PRs), with only 2 patients (1.8%) achieving a CR. In the evaluable population comprising patients with ≥1 on-study postbaseline tumor staging, ORR by either criterion was approximately 42%.
Stephen V. Liu, MD:
In the PD‑L1‑high subset, ORR was 52.6%, which is a little higher than the 44.8% rate seen with pembrolizumab alone in a comparable population in KEYNOTE‑024,22 but still in the same range as what was seen with other approved single-agent ICI therapies.23,24 As we know, patients with newly diagnosed locally advanced or metastatic NSCLC with PD-L1 ≥50% do quite well with IO. In the PD‑L1–low group, the response rate was 41.7%.
The data in the PD‑L1–negative group, where pembrolizumab monotherapy is not an FDA-approved therapy, are what I thought were most interesting. In these patients, the ORR with the combination of pembrolizumab plus eftilagimod alpha was 28.1%. This is encouraging and could provide a new option for patients with PD‑L1–negative tumors, which, as we have noted, represents approximately one third of patients with advanced NSCLC. This is also comparable to what was seen in CheckMate 227 where treatment of patients with PD-L1–negative disease with nivolumab plus ipilimumab resulted in an ORR of 27.3%.25 However, that combination is also not an FDA-approved option for this group.
Stephen V. Liu, MD:
However, durability of response in the PD‑L1–negative group was modest in TACTI-002.21 The median PFS in the intention-to-treat population was 6.9% with a 6-month PFS rate of 55.4%. The median PFS was 11.8 months in the PD‑L1–high subset, 9.3 months in the PD-L1–low group, and 4.2 months in the PD‑L1–negative group. Nonetheless, I would be interested in seeing more data with this drug combination, especially in that PD‑L1–negative setting, which remains an area of unmet need.
Stephen V. Liu, MD:
The most common AEs in the TACTI-002 trial were dyspnea (34.2%, including approximately 11% at grade 3/4, and 1 grade 5 event), asthenia (30.7%), decreased appetite (23.7%), and cough (23.7%).21 Overall, most AEs were grade 1/2 in severity. Eleven patients (9.6%) discontinued treatment due to TRAEs.
Stephen V. Liu, MD:
When looking at any-grade AEs with possible immune etiology, diarrhea (15.8%), hypothyroidism (8.8%), and hyperthyroidism (5.3%) were most common. Four (3.5%) patients developed pneumonitis, and this included 1 grade 4 event and 1 at grade 5.21 These numbers are relatively small, but it will be important to continue monitoring for pneumonitis as eftilagimod continues in development.
Stephen V. Liu, MD:
Eftilagimod alpha plus pembrolizumab is one of many different immunomodulatory combinations under evaluation as we continue the search for ICIs that can further increase response rates in patients with advanced NSCLC. For me, an area of significant unmet need is PD‑L1–negative NSCLC where chemotherapy is either not possible or less appealing in certain populations.
Zofia Piotrowska, MD, MHS:
I agree; it would be really nice to have a chemotherapy‑sparing approach for this group or at least one that might spare chemotherapy until a later line of treatment. I think these are exciting data—not practice changing quite yet but something to keep an eye on going forward.
Stephen V. Liu, MD:
The last study we will discuss regarding use of IO in the advanced setting is a substudy of the Lung‑MAP protocol, a program evaluating novel therapies in patients with stage IV or recurrent, previously treated NSCLC.
Substudy S1800A was a randomized phase II trial of pembrolizumab plus the anti-VEGFR2 monoclonal antibody ramucirumab vs investigator’s choice of standard chemotherapy.26 This study included patients who had already received a PD‑1 or PD‑L1 inhibitor and platinum-based doublet chemotherapy, either together or in sequence, and had to be on therapy at least 84 days before progression. The primary endpoint was OS. Secondary endpoints included ORR, DoR, investigator-assessed PFS, and toxicity.
In substudy S1800A, pembrolizumab plus ramucirumab achieved a median OS of 14.5 months compared with 11.6 months for chemotherapy alone (HR: 0.69; P = .05). Given that most patients (45 of 67) in the standard-of-care arm received docetaxel plus ramucirumab, this study is largely a comparison of docetaxel/ramucirumab to pembrolizumab/ramucirumab, with pembrolizumab/ramucirumab offering a survival advantage.
Of interest, there was no difference in median PFS between the pembrolizumab and standard-of-care arms (4.5 vs 5.2 months) or in ORR (22% vs 28%). Although we do see an OS benefit, there are a number of caveats with this relatively small trial, and it does raise several questions. Is there a relationship between targeting PD‑1 and VEGF? Is there some value in giving VEGF with IO after prior IO? Is this a strategy that should be explored further? Clearly, we will need to wait for more data.
Zofia Piotrowska, MD, MHS:
I agree; it is intriguing. Although docetaxel/ramucirumab is an option in this setting, it is associated with quite significant AEs and is not the easiest regimen to give. Pembrolizumab/ramucirumab is an appealing idea, but this is a small study, and at this point, I would consider the data interesting but not yet ready for incorporating into practice.
I think the question of how to use IO for patients who progress after our current standard-of-care first‑line strategies, whether that be chemotherapy/IO or IO alone, is an open one. There are a lot of ongoing studies that we hope will define whether patients who progress on IO should continue the IO when switching, for example, or add platinum/pemetrexed if they have not yet had that regimen. The INSIGNIA study is looking at that question and will be really important to answering this question (NCT03793179). Other studies are trying to assess whether we could continue with IO and add in other agents to maximize benefit. Maybe there will be biomarkers in the future as well. I think that is another piece of evidence that we need to explore further—are there specific populations where this may be most appealing?
Back
