A Long-Term Survival Picture Emerges as Follow-up Data From NSCLC Trials Mature

Dramatic improvements in efficacy outcomes for patients with non–small cell lung cancer (NSCLC) have been a reason for celebration over the last decade, as FDA approvals of immunotherapy-plus-chemotherapy combinations as well as oncogenic-targeting drugs have transformed the treatment landscape.

The FDA frequently issues approvals for drugs and biologics based on surrogate end points such as progression- free survival (PFS) or objective response rate (ORR) instead of overall survival (OS). These end points are employed for the purposes of expediting drug approvals so that patients with metastatic cancer can receive the newest, cutting-edge therapies. But with the FDA requiring confirmatory trials in many cases to validate data leading to regulatory actions, OS remains the ultimate benchmark of efficacy.

Recently, investigators have released 3-, 4-, and 5-year OS data for key pivotal trials in NSCLC, and the news is good. Updates from 6 clinical trials further emphasize how beneficial these therapies, now established as standard-of-care agents, have been for patients with advanced NSCLC.

“We have had a sense in our clinics for a while that patients with targeted therapies are clearly living longer,” Edward B. Garon, MD, MS, said in an interview with Targeted Therapies in Oncology. “It had been difficult to exactly quantify, but we have now seen in multiple datasets that there are significant improvements in survival in lung cancer, and that at least some of that improvement is based on better treatments for advanced-stage disease.”

A look at these data reveal continuous positive changes in the NSCLC space.

Updates in Immunotherapy

KEYNOTE-001

When investigators published 5-year OS data from the phase 1 KEYNOTE-001 trial (NCT01295827) in June 2019, the results represented the longest follow-up for single-agent pembrolizumab (Keytruda) in advanced or metastatic NSCLC. Total enrollment in KEYNOTE-001 consisted of 550 patients with locally advanced or metastatic NSCLC, including 101 treatment-naive patients and 449 pretreated patients.¹

At data cutoff in November 2018, median follow-up in this trial was 60.6 months. The median OS was 22.3 months (95% CI, 17.1-32.3) in treatment-naive patients and 10.5 months (95% CI, 8.6-13.2) in those who were previously treated. Corresponding estimated 5-year OS rates were 23.2% and 15.5%. In patients with a PD-L1 tumor proportion score (TPS) of 50% or more the 5-year OS rates were 29.6% and 25.0%, respectively.

“That 5-year OS is unprecedented in any kind of metastatic lung cancer trial,” said Andreas Saltos, MD, of the University of South Florida and Moffitt Cancer Center in Tampa. “These data speak to what we tell patients. We see patients every day now who were diagnosed with stage IV lung cancer years ago and are still doing well.”

The KEYNOTE-001 5-year follow-up study notes that 450 patients (82%) in the original cohort had died as of late 2018. “The fact that any patients who were in the early KEYNOTE and CheckMate trials are going on their seventh or eighth year of survival is very encouraging to see,” Saltos said.

KEYNOTE-024

In September, investigators presented 5-year OS data from the KEYNOTE-024 trial (NCT02142738) at the European Society for Medical Oncology (ESMO) Virtual Congress 2020.² The phase 3 trial, which randomized 305 patients to receive either pembrolizumab (n = 154) or physician’s choice standard-ofcare chemotherapy (n = 151), earned pembrolizumab FDA approval as a first-line treatment for patients with PD-L1–positive NSCLC in 2016.³

Patients treated on the trial had previously untreated metastatic NSCLC with PD-L1 TPS of 50% or greater and no targetable EGFR or ALK alterations. At data cutoff in June 2020, the median follow-up was 59.9 months. In the pembrolizumab arm, the median OS was 26.3 months versus 13.4 months in the chemotherapy arm (HR, 0.62; 95% CI, 0.48-0.81). The Kaplan-Meier estimate of the 5-year OS rate was 31.9% in the pembrolizumab group and 16.3% in the chemotherapy group.

Trial protocol permitted patients in the chemotherapy arm to cross over to receive pembrolizumab upon disease progression. The authors reported that 83 patients, or slightly more than half the cohort randomized to chemotherapy, later crossed over to pembrolizumab. Patients who completed 2 years of pembrolizumab therapy also were permitted to resume treatment with pembrolizumab if their disease later progressed.

In total, 39 patients completed the planned 2 years of therapy. At data cutoff, 32 of these patients were still alive with progressive disease leading to death in the remaining 7 patients. The ORR was 82.1% in this patient subgroup, which included 4 patients (10.3%) who experienced a complete response and 28 (71.8%) who had a partial response. Six patients (15.4%) experienced stable disease. Investigators determined that 18 of the patients still alive had not experienced progressive disease at the time of data cutoff.

“Of new data that were recently presented, perhaps the most substantial in terms of informing the community were the 5-year survival data from KEYNOTE-024,” Garon said. “In the original KEYNOTE-001 study, the treatment-na.ve population had a 5-year survival [rate] that was approximately 30%. Now, KEYNOTE-024 confirms that about 30% of these patients with high staining for PD-L1 will be alive 5 years after [starting immunotherapy with single-agent pembrolizumab].”

POPLAR/OAK

Also presented at ESMO 2020, the phase 2 POPLAR (NCT01903993) and the phase 3 OAK (NCT02008227) trial investigators presented 4-year OS data for both trials, which compared the anti–PD-L1 cancer immunotherapy atezolizumab (Tecentriq) with docetaxel in patients with pretreated NSCLC.4 Both randomized, open-label trials formed the basis for the FDA approval of atezolizumab for the treatment of patients with metastatic NSCLC whose disease progressed during or following platinum-containing chemotherapy.⁵

The 2 studies enrolled a total of 1512 patients with locally advanced or metastatic NSCLC who had experienced disease progression during or after prior platinum-based therapy (POPLAR, n = 287; OAK, n = 1225). Patients were randomized to receive either atezolizumab or docetaxel until disease progression. Median follow-up duration was 48.6 months on the POPLAR trial and 47.7 in OAK.

In POPLAR, the 4-year OS rate was 15% in the atezolizumab group with a median OS of 12.6 months. This was compared against docetaxel, with a corresponding OS rate of 8% and a median OS of 9.7 months (HR, 0.76; 95% CI, 0.58-1.00; P = .0455).

In OAK, patients in the atezolizumab group experienced a median OS of 13.3 months and a 4-year OS rate of 16%. In the docetaxel group, corresponding data were 9% and 9.8 months (HR, 0.78; 95% CI, 0.68-0.89; P < .0001).

Saltos noted that the OS results from POPLAR and OAK results are comparable to those observed in the pretreated pembrolizumab arms of KEYNOTE-001. “The higher the PD-L1 level, the higher the chance of a response which is one aspect to consider when comparing these trials to each other to help guide our clinical decision-making process,” he said. “Importantly, while PD-L1 is a good biomarker prognostically, the needle is still moving for patients without PD-L1 expression to have a better chance of long-term survival.”

PACIFIC

The double-blinded phase 3 PACIFIC trial (NCT02125461) randomized patients with unresectable, stage III NSCLC to durvalumab (Imfinzi) or placebo following chemoradiotherapy, regardless of their PD-L1 status.6 Patients disease could not have progressed following concurrent platinum-based chemotherapy plus radiation. These data led to FDA approval of durvalumab for this indication in NSCLC.⁷

Of 713 patients, 476 received durvalumab and 237 received placebo. As of the March 2020 data cutoff, the median follow-up was 34.2 months. The key results remained consistent with earlier reports, which showed that both PFS (stratified HR, 0.55; 95% CI, 0.44-0.67) and OS (stratified HR, 0.71; 95% CI, 0.57-0.88) favored the durvalumab arm.

The median OS of the durvalumab arm was announced during ESMO 2020, at 47.5 months compared with 29.1 months for placebo. The 48-month OS rates were 49.6% for durvalumab and 36.3% for placebo. PFS rates at 48 months were 35.3% versus 19.5%, respectively. Updated patient subgroup treatment effects will be reported in the future.

“PACIFIC’s maximum treatment course was 1 year, but these responses have been sustained for at least 3 years now, suggesting that a higher number of these patients may be cured of their disease,” Saltos said. “These data show about half of patients treated with durvalumab survived 4 years, and an estimated 35% had not progressed, a significant advance.”

Targeted Therapy Updates

Osimertinib

Earlier this year, investigators published long-term OS data for the FLAURA trial (NCT02296125) in the New England Journal of Medicine.⁸ This phase 3 trial compared firstline osimertinib (Tagrisso) with standard-of-care EGFR tyrosine kinase inhibitors (TKIs) in patients with EGFR mutation–positive advanced NSCLC. These data served as the basis of the FDA’s approval of osimertinib for first-line treatment of metastatic NSCLC with most common EGFR mutations.⁹

In total, 556 patients with previously untreated advanced NSCLC with an EGFR mutation (exon 19 deletion or L858R allele) were randomized to receive either osimertinib or treatment with a first-generation EGFR TKI, either gefitinib (Iressa) or erlotinib (Tarceva). In the osimertinib group, the median OS was 38.6 months and 31.8 months in the comparator group (HR, 0.80; 95.05% CI, 0.64-1.00; P = .046).

The median exposure was 20.7 months in the osimertinib group and 11.5 months in the comparator group. Nearly 30% of osimertinib treated patients (n = 79, 28%) continued to receive therapy at 3 years, while only 26 (9%) in the comparator group did.

The new FLAURA data confirm an impression that Garon has held regarding well-tolerated, effective therapy. “From a progression-free survival perspective, the events were reasonably mature previous to overall survival. Fortunately, those results have lagged because this population has seen substantial outcome improvements,” he said. “The data from FLAURA have clearly changed practice, and we continue to see an improvement in outcomes for patients with these targets.”

Alectinib

Updated OS results for the ALEX trial (NCT02075840), which compared alectinib (Alecensa) with crizotinib (Xalkori) in patients with treatment-na.ve ALK-positive NSCLC, appeared in the Annals of Oncology earlier this year.10 ALEX was the first global randomized study to show clinically meaningful improvement in OS for a next-generation ALK TKI versus crizotinib in the treatment of patients with ALK-positive NSCLC who had no prior therapies. These data served as the basis of the 2017 FDA approval of alectinib as frontline therapy for ALK-positive NSCLC.¹¹

ALEX randomized 303 patients with stage III or IV ALK-positive NSCLC to receive alectinib (n = 152) or crizotinib (n = 151) until disease progression, toxicity, withdrawal, or death. Investigator-assessed PFS was the trial’s primary end point.

At the time of the 5-year OS data cutoff in November 2019, the median duration of OS follow-up was 48.2 months in the alectinib arm versus 23.3 months in the crizotinib arm. OS data remain immature with 37% of survival events occurring as of data cutoff; the target maturity for survival is 50%, per the trial protocol. The median OS was not reached with alectinib versus 57.4 months with crizotinib (stratified HR, 0.67; 95% CI, 0.46-0.98.

With alectinib, the 5-year OS rate was 62.5% versus 45.5% with crizotinib. Corresponding median treatment duration was 28.1 months and 10.8 months. At data cutoff, 34.9% of patients in the alectinib group and 8.6% of patients in the crizotinib group were still on study treatment.

Patients who had central nervous system metastases at baseline saw an OS benefit from alectinib (HR, 0.58; 95% CI, 0.35-1.00), which Saltos pronounced a “significant development” in this population. He said “this recent report adds weight to the idea that alectinib up front is superior” to other available agents in this patient population.

Indeed, the study authors noted that the use of next-generation ALK TKIs as salvage therapy is “feasible,” but their impact on OS in this population remains uncertain.

Looking Ahead in NSCLC

Both Saltos and Garon look ahead to the next round of trial results for these and newer agents with excitement, noting that initial concerns about using genomic data to drive treatment decisions beyond cases of EGFR, ALK, and ROS1 mutation are being laid to rest.

“In the past several months, we have seen approvals for gene rearrangements involving the [capmatinib (Tabrecta)] approval for the MET exon 14 skipping mutation,12 which is a fairly common mutation,” Garon said. “Also on the horizon are multiple agents targeting KRAS G12C mutations, which we think are going to turn out to be a group very similar in size to the patients with EGFR positivity, particularly in Western populations.”

As intriguing as future therapeutics are in NSCLC, it’s also natural to reflect on how far the profession has come in a relatively short time.

“I’m having different conversations with my patients these days. A lot of the patient information on the web still refers to the average survival rate as 1 year in advanced NSCLC,” Saltos said. “It can take a lot of reassurance to help patients understand how much the landscape has changed. But we all know that we’re seeing many patients live longer than a year thanks to these new therapies (TABLE^2,4,6,8,10).”

_References:

_{1. Garon EB, Hellmann MD, Rizvi NA, et al. Five-year overall survival for patients with advanced non‒small-cell lung cancer treated with pembrolizumab: results from the phase I KEYNOTE-001 study. J Clin Oncol. 2019;37(28):2518-2527. doi:10.1200/JCO.19.00934}

_{2. Brahmer JR, Rodriguez-Abreu D, Robinson AG, et al. KEYNOTE-024 5-year OS update: first-line (1L) pembrolizumab (pembro) vs platinum- based chemotherapy (chemo) in patients (pts) with metastatic NSCLC and PD-L1 tumour proportion score (TPS) ≥50%. Ann Oncol. 2020;31 (suppl 4):S1142-S1215. Abstract LBA51. doi:10.1016/j.annonc. 2020.08.2284}

_{3. Pembrolizumab (Keytruda) checkpoint inhibitor. News release. FDA. October 24, 2016. Accessed October 18, 2020. https://bit.ly/37qP8Wc}

_{4. Mazieres J, Rittmeyer A, Gadgeel SM, et al. 4-year survival in randomised phase II (POPLAR) and phase III (OAK) studies of atezolizumab vs docetaxel in pre-treated NSCLC. Ann Oncol. 2020;31(suppl 4):S1142-S1215. Abstract 1271P. doi:10.1016/j.annonc.2020.08.1585}

_{5. Atezolizumab (Tecentriq). News release. FDA. October 19, 2016. Accessed October 18, 2020. https://bit.ly/2FLXxIx}

_{6. Faivre-Finn C, Vicente D, Kurata T, et al. Durvalumab after chemoradiotherapy in stage III NSCLC: 4-year survival update from the phase III PACIFIC trial. Ann Oncol. 2020;31(suppl 4):S1142-S1215. doi:10.1016/annonc/annonc325}

_{7. FDA approves durvalumab after chemoradiation for unresectable stage III NSCLC. News release. FDA. February 20, 2018. Accessed October 19, 2020. https://bit.ly/37oQGAb}

_{8. Ramalingam SS, Vansteenkiste J, Planchard D, et al. Overall survival with osimertinib in untreated, EGFR-mutated advanced NSCLC. N Engl J Med. 2020;382(1):41-50. doi:10.1056/NEJMoa1913662}

_{9. FDA approves osimertinib for first-line treatment of metastatic NSCLC with most common EGFR mutations. News release. FDA. April 19, 2018. Accessed October 19, 2020. https://bit.ly/3o9BEUP}

_{10. Mok T, Camidge DR, Gadgeel SM, et al. Updated overall survival and final progression-free survival data for patients with treatment-naive advanced ALK-positive non-small-cell lung cancer in the ALEX study. Ann Oncol. 2020;31(8):1056-1064. doi:10.1016/j.annonc.2020.04.478}

_{11. Alectinib approved for (ALK) positive metastatic non-small cell lung cancer (NSCLC). News release. FDA. November 7, 2017. Accessed October 19, 2020. https://bit.ly/2T9IXxB}

_{12. FDA grants accelerated approval to capmatinib for metastatic nonsmallcell lung cancer. News release. FDA. May 6, 2020. Accessed October 19, 2020. https://bit.ly/2HkEANq}