乌鸦传媒

Five is the "magic" number in the CheckMate-017 and CheckMate-07 trials: 5-year outcomes data from the randomized phase III trials of nivolumab (Opdivo) in previously treated, advanced non-small cell lung cancer (NSCLC) showed a five-fold increase in the overall survival (OS) rate, with no new safety signals.

Specifically, 5-year pooled OS rates were 13.4% for the nivolumab trials vs 2.6% for docetaxel, while 5-year progression-free survival (PFS) rates were 8.0% vs 0%, respectively, according to Hossein Borghaei, DO, of Fox Chase Cancer Center in Philadelphia, and colleagues.

In addition, as they reported in the , nivolumab-treated patients without disease progression at 2 and 3 years had an 82.0% and 93.0% chance of survival, respectively, and a 59.6% and 78.3% chance of remaining progression-free at 5 years, respectively.

Finally, treatment-related adverse events (TRAEs) were reported in 25.8% of the patients on nivolumab at 3 to 5 years of follow-up, 3.2% of whom experienced new events (grade 3 increased lipase).

"Historically, 5-year survival rates of patients with advanced [NSCLC] who received chemotherapy were <5%," the authors wrote. "Among ≥5-year survivors in the nivolumab arm ... the median duration of therapy was 36.9 months, and 18 of 50 remained on nivolumab at 5 years, suggesting that some patients may achieve long-term survival with continuous nivolumab treatment ... These findings represent an important advancement in the treatment of lung cancer."

Nivolumab also proved its mettle in other NSCLC settings, which were reported at the 2020 European Society for Medical Oncology virtual meeting from researchers in Korea and Germany, as well as at the ASCO 2020 virtual annual meeting by investigators from Germany.

The following Q&As cover the researchers' main findings:

What was the impetus for the ""?

Jong-Seok Lee, MD, Seoul National University Bundang Hospital: Chemotherapy and angiogenesis inhibitors are expected to have a synergistic effect in combo with immune checkpoint inhibitors. Chemotherapy activates the anti-tumor response by releasing immunogenic tumor antigens, inducing T-cell priming and recruitment of T cells.

In addition, angiogenesis inhibitors have been been shown to inhibit immune evasion of cancer cells by normalizing the tumor microenvironment.

A pilot trial [] showed tolerable safety profiles and a promising efficacy of nivolumab with carboplatin, paclitaxel, and bevacizumab (Avastin) combination in patients with advanced NSCLC. With this background, we conducted the phase III clinical ONO-52 trial [ONO-4358-52/TASUKI-52 trial].

What were some of the main findings?

PFS was the primary endpoint. Nivolumab with chemotherapy and bevacizumab significantly improved PFS as compared to placebo with chemotherapy and bevacizumab.

The median PFS was 12 months in the nivolumab arm and about 8 months in the placebo arm. The HR was 0.06 [0.56] with a P value of <0.0001 ... the Kaplan Meir curve is clearly separated at all stages.

The 12-month PFS rate was 50% in the nivolumab arm and approximately 30% in the placebo arm. These results clearly demonstrated that nivolumab in combination with carboplatin, paclitaxel, and bevacizumab therapy has excellent clinical mean increase. Also, PFS was prolonged regardless of PD-L1 expression levels.

We also performed PFS analysis in key subgroups, and found consistent PFS benefit in the nivolumab group across almost all of the subgroups. The overall response rate [ORR] was 61% in the nivolumab group and 50.5% in the placebo arm.

The number of patients with serious AEs, and with AEs leading to treatment discontinuation, was numerically higher in the nivolumab arm compared with the placebo arm. However, TRAEs leading to dose delay and death were similar in both arms [48.84% vs 44.7% and 1.8% vs 1.5%, respectively].

TRAEs with an instance of 20% or more were chemotherapy- or bevacizumab-induced events and, generally, the instance appeared not to be different between the treatment arms. Rash was reported in the nivolumab arm more than in the placebo arm -- 29.7 vs 14.5 for any grade TRAE. However, rash is a well-known AE with nivolumab.

In this interim analysis for OS, the data is not mature [as of Sept. 2020], so we need long-term follow-up. Regarding safety, no new safety signals were found in the nivolumab arm.

What was the impetus for the ""?

Farastuk Bozorgmehr, MD, Thoraxklinik at Heidelberg University Hospital: The hypothesis of the FORCE study was based on the so-called abscopal effect, namely the enhancement of an anti-tumor immunity, which would then be boosted by immune checkpoint blockade.

The study was designed to include patients with a clinical need of radiotherapy of a metastatic site. Patients would be in second- or third-line treatment and would then be included to receive 20 Gy radiotherapy in five fractions with concurrent nivolumab.

Sample-size justification, and the primary endpoint, was based on the hypothesis that we would be able to achieve an ORR of higher than 19% with patient numbers of 65 in each group.

To allow for biomarker acquisition, four times per patient, in patients without radiotherapy, and to be better able to clinically characterize these patient groups, we decided to include the same number of patients in another group receiving only nivolumab.

We were able to recruit 101 patients: 41 patients receiving radiotherapy and nivolumab (group A), and 60 receiving nivolumab only (group B]). Due to changes in the first-line treatment with the approval of pembrolizumab [Keytruda] in combination with chemo in the treatment of NSCLC, the recruitment [for FORCE] dropped over time, and in December 2019 we decided to halt recruitment. We had already achieved the patient number that would allow for descriptive analysis of our primary endpoint.

What were some of the main findings?

Patients with unfavorable characteristics were more prevalent in the group receiving nivolumab plus radiotherapy: more patients with ECOG performance status of 1 were included in this group -- more patients with a higher tumor load and more patents with extrathoracic metastases. Within these patients with extrathoracic metastases, there were more with bone metastases who have been shown to respond poorly to immunotherapy in retrospective analysis.

With these baseline characteristics in mind, the primary objective was to achieve an ORR higher than 19% in group A, which we could not achieve (P=0.0991 for one-side binomial test). In group A, the imputed RR was 8% and in group B, the imputed RR was 24%, and this was explained by the better clinical status of patients recruited in the latter group.

The combination of nivolumab and palliative radiotherapy was safe and feasible, but with an imputed ORR of 8%, ORR in group A was well below expected. We hope to gain more insight into this from our secondary endpoints, which are still pending [as of Sept. 2020]. These are biomarker analysis, data on treatment beyond progression, and survival analysis.

What was the impetus for ""?

Martin Reck, MD, PhD, Hospital Grosshansdorf: Nivolumab and ipilimumab (Yervoy) are two distinct but complementary checkpoint inhibitors, and the combination of both has shown long-term overall survival benefit across various tumor entities.

In , an improved efficacy in first-line treatment of NSCLC was observed independent from the PDL1 expression and histology. We hypothesized that ending two cycles of chemotherapy could provide rapid disease control while building on the long-term survival benefit by the combination. We presented the results of the pre-specified interim analysis as well as additional efficacy and safety outcomes with a longer follow-up.

What were some of the main findings?

The interim analysis showed a significant improvement in OS and benefit in favor of the combination arm corresponding to an HR of 0.69. The Kaplan-Meier curves were separating early and consistently without crossing.

PFS and response were tested hierarchically and were also significantly improved in favor of the combination arm. With a further follow-up at 4 months, the confirmation and improvement of OS was observed, now with an HR of 4.66, a median OS of 15.6 compared to 10.9 months, and a 1-year OS rate of 63% compared to 47%.

Of note, 34% of patients in the chemotherapy arm received a subsequent treatment with an immunotherapy following progression of the disease. The benefit in OS was seen in the maturity of investigated subgroups.

There was a reduced efficacy in the small groups of patients older than 75 and never-smoking patients on the other side. There was an interesting signal survival benefit in the group of patients with brain metastases. The magnitude of benefit was seen across histologies, with an HR of 0.69 for patients with a non-squamous NSCLC on the left and a hazard ratio of 0.62 for patients with squamous NSCLC on the right.

Furthermore, the survival benefit was also independent from the PD-L1 expression level, with an HR of 0.62 for patients with PD-L1 negative tumors and 0.64 for patients with PD-L1 positive tumors. In addition, the benefit was consistent across the various levels of PD-L1 expression.

Also, PFS was improved in favor of the combination arm, with an HR of 0.68 and a 1-year PFS rate of 33%, compared with 18% in the control arm.

The fatal toxicity rate was low -- 2% in each arm. The most frequent TRAEs were nausea, anemia, asthenia, and diarrhea.

So in summary, CheckMate 9LA met its primary endpoint of OS at the pre-planned interim analysis with an HR of 0.69 and a P value of 0.0006. The clinically meaningful improvement of oval efficacy endpoints was observed and increased with longer follow-up.

The magnitude of benefit with the nivolumab and ipilimumab, plus two cycles of chemotherapy was consistent across all histologies and PD-L1 expression levels, including the PD-L1 negative tumors. No new safety signals were observed for the combination.

And with an early separation of the survival curves and the lower primary progression rates, the hypothesis for the CheckMate 9LA study design was validated.

Read the study here and expert commentary about the clinical implications here.