Kelly Explores Current Immunotherapy/Chemotherapy Combinations in NSCLC

Ronan J. Kelly, MD, MBA

Chief of Oncology

Charles A. Sammons Cancer Center

W.W. Caruth Jr Chair in Immunology

Baylor University Medical Center

Dallas, TX

CASE SUMMARY

• A 62-year-old woman presented to the emergency department with vague complaints of voice changes and cough. She had recent weight loss of 11 lb.
• CT of thorax discovered a 4-cm nodule in left upper lobe.
• CT of abdomen also revealed metastases to the liver.
• MRI of brain was negative for brain metastases.
• Past medical history
  • Hypertension, hyperlipidemia, chronic obstructive pulmonary disease (managed on inhalers)
  • Smoking: 10 to 15 pack-years, stopped 25 years ago
• Family history
  • Mother died from lung cancer at age 65 years.
    
• Physical exam
  • Weight: 125 lb
  • ECOG performance status: 1
    
• Final pathology consistent with squamous cell carcinoma; metastatic stage IV
• PD-L1 expression by immunohistochemistry was 0%.
• Next-generation sequencing: no actionable mutations
• Treatment options were reviewed with the patient and family.
• Patient was initiated on nivolumab (Opdivo) 360 mg intravenously (IV) every 3 weeks plus ipilimumab (Yervoy) 1 mg/kg IV every 6 weeks plus 2 cycles of chemotherapy every 3 weeks.

TARGETED ONCOLOGY: What data support the use of nivolumab/ ipilimumab plus chemotherapy in patients with previously untreated stage IV non–small cell lung cancer (NSCLC)?

KELLY: In the CheckMate 9LA study [NCT03215706], the question that the investigators were looking to answer was [whether] they could maximize cell kill at the outset with a dose-intensive regimen straight at the start, which was 2 cycles of chemotherapy combined with IO [immuno-oncology].

We had a lot of data on IO use alone, but the question they wanted to ask was [whether] we can cause cytotoxic cell death with the chemotherapy agents, [such as] pemetrexed [Alimta], cisplatin [Platinol], or carboplatin [Paraplatin] and paclitaxel [Taxol]. Can you cause cell death [that] causes neoantigen release and changes the microenvironment, then add an IO doublet?

It was only ever meant to be 2 cycles, because they wanted to see [whether] you could take away the chemotherapy and have the patient on doublet IO for less toxicity and a little better quality of life.

The comparator arm was 4 cycles of chemotherapy given every 3 weeks. As is normal, the standard maintenance was up to 2 years or until unacceptable toxicity. The key eligibility criteria were stage IV or recurrent [NSCLC], no prior treatments, no EGFR or ALK alterations, and good ECOG performance status. They were stratified by sex, PD-L1 status, or histology [squamous or nonsquamous]. The primary end point was overall survival [OS], and the secondary end points were progression-free survival [PFS], overall response rate [ORR], and efficacy by tumor PD-L1 expression. Safety was an exploratory end point.¹

Based on the 3-year data update, the median OS was 15.8 months for the IO doublet arm vs 11 months for chemotherapy alone [HR, 0.74; 95% CI, 0.62-0.87]. Quite impressively, 27% of the patients were still alive at 3 years.² I know most of us, when we were training, struggled to get a 1-year OS in lung cancer. The 3-year OS is [approximately] 30%. Will we see more [patients] doing well in years 4 and 5? We will have to wait and see, but 27% at 3 years is quite promising. In terms of the other efficacy end points, the PFS was 6.4 months for the IO arm vs 5.3 months for the chemotherapy-alone arm [HR, 0.70; 95% CI, 0.59-0.83]. The ORR was 38% for the IO arm vs 25% for the chemotherapy-alone arm. The duration of response [DOR] was also longer for the IO arm, [at 12.4 months vs 5.6 months for the chemotherapy arm]. The ongoing response at 3 years was 23% for the IO arm vs 14% for the chemotherapy arm.²

What I find fascinating in this study is the subgroup analysis. We always must be careful because they are not always powered for this, but a couple subgroups are quite interesting to me. Did PD-L1 expression matter? [Approximately] one-third of patients have a PD-L1 of less than 1%, one-third of them have a PD-L1 of 1% to 49%, and then [approximately] the remaining one-third have a PD-L1 of greater than 50%. That means [approximately] one-third of our patients are low expressors. Did that matter? No, not on this study. There was an efficacy improvement in the median OS regardless of that. For patients with a PD-L1 [expression] greater than 1%, it was 15.8 months for IO doublet vs 10.9 months for chemotherapy [HR, 0.74; 95% CI, 0.60-0.93]. For patients with a PD-L1 of less than 1%, it was 17.7 months vs 9.8 months, respectively [HR, 0.67; 95% CI, 0.51-0.88]. For patients with a PD-L1 greater than 1%, 28% of them were alive at 3 years. For [those with a] PD-L1 [of] less than 1%, 25% were alive. So PD-L1 [expression] doesn’t seem to matter. We have seen it in other studies when we use a CTLA-4 inhibitor.

What is that CTLA-4 inhibitor doing that helps overcome those low PD-L1 patient scores? We know the CTLA-4 inhibitor helps generate more cytotoxic T cells, whereas the PD-1 inhibitors may not work as well if you don’t have the T cells in place. That is one hypothesis. If you look at the [group with a] PD-L1 of greater than 50%…we know that group has done very well historically. The median OS for that group in this subgroup analysis was 18.9 months vs 12.9 months [HR, 0.75, 95% CI, 0.53-1.07].

What was the significance of the results for patients with brain metastases in CheckMate 9LA?

I talk to a lot of my colleagues at Johns Hopkins Medicine, [The University of Texas] MD Anderson [Cancer Center], and Dana-Farber [Cancer Institute] about the brain metastases data [Figure³]. These were all treated [brain metastases]. I don’t remember seeing HRs like this in lung cancer in a nononcogenic-driven tumor. The median OS in the brain metastases group was 19.3 months [for chemoimmunotherapy] vs 6.8 months [for chemotherapy], with an HR of 0.43 [95% CI, 0.27-0.67]. This is fascinating. The OS rate at 24 months in the brain metastases group was 35% vs 12%, respectively.³

I don’t think we understand the biology here. It’s not so easy to take [cerebrospinal fluid] and understand [intracranial antitumor activity], and [we are not] doing brain biopsies. We will tease this out, but it just shows that something is happening where you are generating more cytotoxic T cells, and I believe they are penetrating the blood-brain barrier.

Around the United States, more and more lung cancer specialists are leaning [toward] the CTLA-4 inhibitor in patients with brain metastases. We must be careful because it is subgroup analysis, so [we should] look at whether it is powered for this. But we’ve seen this in the CheckMate 277 trial [NCT02477826]—the other nivolumab plus ipilimumab study—where the brain metastases issue arose as well.⁴ I think we are beginning to see that the combination IOs seem to be helping patients with brain metastases. We will have to see more data, but it is certainly promising. For years, we’ve been thinking the brain is the sanctuary site and we couldn’t get our treatments in there.

Chemotherapy alone is probably not going to do too well. Will chemotherapy plus a PD-1 inhibitor do so? That wasn’t [investigated in] the trial design. In the CheckMate 9LA study, patients without brain metastases had a median OS of 15.6 months for the IO arm vs 12.1 months [with] chemotherapy [HR, 0.79; 95% CI, 0.65-0.95].³

How do you handle dosing and toxicity for this combination regimen?

I also do esophagogastric oncology, and I led a study for gastric cancer internationally with durvalumab [Imfinzi] plus tremelimumab-actl [Imjudo]. We were giving it too high and too frequently, and we had colitis and a lot of toxicity. The CheckMate 9LA study probably had that similar issue at the start with nivolumab and ipilimumab.… Ten or 12 years ago…a lot of the physicians in the community were a little nervous about the combination.

They thought this would cause a lot of toxicity. What we are seeing now is [that] the lower dose and less frequent administration is better tolerated. Will there be IO toxicities? Yes. Are they higher? Yes. But we all know how to manage them.

The million-dollar question is: Which one is causing the toxicity? And if you want to stop one, which one do you stop? No one is going to definitively be able to answer that. I normally stop the CTLA-4 inhibitor, but [physicians] will be able to make their own decisions. I [stop the CTLA-4 inhibitor] because I am more used to the PD-1 inhibitors, but it is something to think about. Toxicities included skin, endocrine, [gastrointestinal], hepatic, and renal reactions. The take-home was [there were] no major toxicity signals that were worrisome. It was quite well tolerated for those patients.¹

There were some grades 3 and 4 hematologic toxicities. The one we always [must] watch out for is pneumonitis. I have lost patients with pneumonitis, and patients have been referred to me where pneumonitis had been missed. We should always have it at the top of our list, because it can really catch us if we don’t watch out for it.

Patients who discontinued treatment due to a treatment-related adverse event still had an improvement, so even if you get these treatments and then stop because of the toxicity, patients may benefit. Of course, you must get them through the toxicity, but if you don’t let it get to a grade 3 or 4 colitis or diarrhea, and maybe you recognize it earlier, there may be a potential benefit for the patient. [However], this is also a small group of only 61 patients.⁵ Take this with a little bit of intellectual curiosity.

Which trial investigated pembrolizumab (Keytruda) plus chemotherapy in patients with untreated squamous stage IV NSCLC?

The KEYNOTE-407 trial [NCT02775435] was an older study of pembrolizumab plus chemotherapy vs chemotherapy alone in untreated stage IV NSCLC with squamous histology. They excluded patients with symptomatic brain metastases and pneumonitis requiring steroids. The stratification factors were PD-L1 status, choice of taxane [paclitaxel or nab-paclitaxel], and geographic region. Patients were randomly assigned to pembrolizumab, carboplatin, and a taxane vs placebo plus the same chemotherapy. They did maintenance pembrolizumab or placebo for up to 31 cycles. The primary end point was PFS, and secondary end points were ORR, DOR, and safety.

The 5-year updated data were presented at the [European Society for Medical Oncology Congress 2022]. The 5-year OS rate was 18.4% [for the] pembrolizumab arm vs 9.7% for the placebo arm. The OS by PD-L1 status was also reported, with a HR of 0.68 for those with PD-L1 greater than 50%, 0.61 for those with a PD-L1 ranging from 1% to 49%, and 0.83 for those with a PD-L1 of less than 1%.⁶ Physicians have choices now when it comes to lung cancer. We often use the older regimen, the one we are more used to. But for me, in the group with PD-L1 of less than 1%, I am thinking of a CTLA-4 IO, [such as] in the CheckMate 9LA study in patients with brain metastases. That is something I am factoring in. But [physicians] all have their own treatment decisions to make.

[In terms of] toxicities, this is a very old study that we are…very familiar with, but nothing was jumping out at us. There were some immune-related AEs. The rate of pneumonitis was 6.5% vs 2.1% on chemotherapy alone. Other [notable AEs] were colitis and skin reactions.5

What are some other newer therapies? Do the data support their use?

There is another player on the market…cemiplimab-rwlc [Libtayo]. The EMPOWER-Lung 3 study [NCT03409614] used cemiplimab plus a platinum doublet in metastatic NSCLC with any PD-L1 expression and both squamous and nonsquamous histology. Maintenance treatment with cemiplimab or placebo was continued for 108 weeks [or] until progression of disease or unacceptable toxicity. The median OS was [21.9 months for the cemiplimab arm vs 13 months for placebo, (HR, 0.71; 95% CI, 0.53-0.93; P = .014)].⁷

[It is a] challenge for all these [therapies] to get traction. We need something to require us to change, [such as] if there is better efficacy. I don’t think we are seeing that with this combination.

The POSEIDON study [NCT03164616] had durvalumab plus tremelimumab, which is a PD-L1 and CTLA-4 inhibitor vs chemotherapy. It was a very large study, with [more than] 1000 patients being randomly assigned on a 1:1:1 basis. Both squamous and nonsquamous histology were included.⁸ I had a gastroesophageal cancer trial exactly like this.

The durvalumab plus or minus tremelimumab and chemotherapy were given every 3 weeks for 4 cycles, then you could discontinue the chemotherapy. The end points were PFS and OS. The dosing for durvalumab was 1500 mg, and the limited course tremelimumab was 75 mg. The PFS met statistical significance [HR, 0.74; 95% CI, 0.62-0.89; P = .0009], but the OS didn’t [HR, 0.86; 95% CI, 0.72-1.02; P = .0758].⁸ I am not sure where that puts durvalumab plus chemotherapy. I haven’t been using it. Physicians can decide [whether] they feel comfortable with it.

The other arm was durvalumab, tremelimumab, and chemotherapy vs chemotherapy alone. The median PFS was 26.6 months vs 13.1 months [HR, 0,72; 95% CI, 0.60-0.86; P = .0003], and the median OS was 14.0 months vs 11.7 months for chemotherapy [HR, 0.77; 95% CI, 0.65-0.92; P = .003]. So it was positive for PFS and for OS. It has been out for [only] a little bit of time, but it is something to think about.

_REFERENCES

_{1. Paz-Ares L, Ciuleanu TE, Cobo M, et al. First-line nivolumab plus ipilimumab combined with two cycles of chemotherapy in patients with non-small-cell lung cancer (CheckMate 9LA): an international, randomised, open-label, phase 3 trial. Lancet Oncol. 2021;22(2):198-211. doi:10.1016/S1470-2045(20)30641-0}

_{2. Paz-Ares LG, Ciuleanu TE, Cobo-Dols M, et al. First-line (1L) nivolumab (NIVO) + ipilimumab (IPI) + 2 cycles of chemotherapy (chemo) versus chemo alone (4 cycles) in patients (pts) with metastatic non-small cell lung cancer (NSCLC): 3-year update from CheckMate 9LA. J Clin Oncol. 2022;40(suppl 17):LBA9026. doi:10.1200/JCO.2022.40.17_suppl.LBA9026}

_{3. Carbone D. First-line nivolumab + ipilimumab + chemo in patients with advanced NSCLC and brain metastases: Results From CheckMate 9LA. Presented at: 2021 World Conference on Lung Cancer; September 8-14, 2021; Virtual. Abstract OA09.01.}

_{4. Borghaei H, Hellmann MD, Paz-Ares LG, et al. Nivolumab (nivo) + platinum-doublet chemotherapy (chemo) vs chemo as first-line (1L) treatment (Tx) for advanced non-small cell lung cancer (NSCLC) with <1% tumor PD-L1 expression: results from CheckMate 227. J Clin Oncol. 2018;36(suppl 15):9001. doi:10.1200/JCO.2018.36.15_suppl.9001}

_{5. Reck M, Ciuleanu TE, Cobo M, et al. First-line nivolumab plus ipilimumab with two cycles of chemotherapy versus chemotherapy alone (four cycles) in advanced non-small-cell lung cancer: CheckMate 9LA 2-year update. ESMO Open. 2021;6(5):100273. doi:10.1016/j.esmoop.2021.100273}

_{6. Novello S, Kowalski DM, Luft A, et al. 5-year update from KEYNOTE-407: pembrolizumab plus chemotherapy in squamous non-small cell lung cancer (NSCLC). Ann Oncol. 2022;33(suppl 7):S993-S994. doi:10.1016/j.annonc.2022.07.1102}

_{7. Gogishvili M, Melkadze T, Makharadze T, et al. Cemiplimab plus chemotherapy versus chemotherapy alone in non-small cell lung cancer: a randomized, controlled, double-blind phase 3 trial. Nat Med. 2022;28(11):2374-2380. doi:10.1038/s41591-022-01977-y}

_{8. Johnson ML, Cho BC, Luft A, et al. Durvalumab with or without tremelimumab in combination with chemotherapy as first-line therapy for metastatic non-small-cell lung cancer: the phase III POSEIDON study. J Clin Oncol. 2023;41(6):1213-1227. doi:10.1200/JCO.22.00975}