Lenvatinib Plus Pembrolizumab Fails to Improve OS and PFS Over Docetaxel in Stage IV NSCLC

Treatment with lenvatinib (Lenvima) plus pembrolizumab (Keytruda) did not provide a survival advantage compared with standard-of-care docetaxel in patients with advanced-stage non–small cell lung cancer (NSCLC) who experienced disease progression following prior exposure to a PD-L1 inhibitor and platinum-based chemotherapy, according to findings from the phase 3 LEAP-008 trial (NCT03976375) presented during the 2023 ESMO Immuno-Oncology Congress.

Results from the final analysis of overall survival (OS) revealed that 81.1% of patients experienced an OS even with lenvatinib plus pembrolizumab compared with 81.5% of patients treated with docetaxel (hazard ratio, 0.98; 95% CI, 0.78-1.23; one-sided P = 0.4342). Moreover, the combination elicited a median OS of 11.3 months (95% CI, 9.4-13.2) compared with 12.0 months (95% CI, 9.6-13.7) among those who received docetaxel. The 12-, 24-, and 36-month OS rates in the combination arm were 46.5%, 21.1%, and 16.8%, respectively; corresponding percentages in the docetaxel arm were 49.7%, 19.8%, and 14.6%.

Similarly, the lenvatinib combination produced PFS events in 82.7% of patients, with a median PFS of 5.6 months (95% CI, 4.2-6.5). Comparatively, the PFS rate was 75.7% in the docetaxel arm, with a median PFS of 4.2 months (95% CI, 3.2-5.2). The hazard ratio favored the combination at 0.89 (95%, 0.70-1.12). The 12-, 24-, and 36-month PFS rates in the combination arm were 17.5%, 7.3%, and 7.3%, respectively; corresponding percentages in the docetaxel arm were 13.5%, 5.2%, and 3.9%.

Moreover, no statistically significant benefit with the combination over docetaxel was observed in any prespecified subgroups.

“We did not find a subgroup that benefited from lenvatinib plus pembrolizumab over docetaxel, although there is some question in this small number of patients that did not receive a checkpoint inhibitor as immediate therapy but had other sequencing,” Leighl added during the presentation,” lead study and presenting author Natasha B. Leighl, MD, BSc, MMSc, a clinician investigator at Princess Margaret Cancer Centre in Toronto, Canada, stated during the meeting.

Despite various combinations being tested in phase 3 studies, none have significantly increased efficacy vs standard docetaxel for patients with stage IV NSCLC who have progressed after prior anti–PD-(L)1 therapy and platinum-based chemotherapy.

The multi-TKI lenvatinib targets VEGFR, FGFR, PDGFR, RET, and KIT, thereby playing a crucial role in immune stimulation of the tumor microenvironment. According to previously reported findings, lenvatinib exhibited antitumor activity when used in combination with carboplatin and paclitaxel as a first-line or subsequent therapy for stage IIIB or IV NSCLC, in combination with best supportive care as a third-line or later therapy for advanced NSCLC, and as a first-line or later monotherapy for RET-positive lung adenocarcinoma.

“The rationale for this study was the exciting activity of [lenvatinib plus pembrolizumab] that we saw in pretreated patients with NSCLC who had [progressed on] immunotherapy and platinum-based treatment,” Leighl explained in her oral presentation. “[These patients] had a response rate to the combination of 33% and a median duration of response of 10.9 months. [The combination also produced] very exciting data in patients with renal cell carcinoma and melanoma whose disease had previously progressed on immunotherapy.”

To address this unmet need for combination therapies that will enhance outcomes in this population, investigators conducted the LEAP-008 study of lenvatinib both as a monotherapy or in combination with pembrolizumab vs docetaxel.

The open-label trial enrolled patients with histologically or cytologically confirmed stage IV NSCLC who experienced disease progression following one anti–PD-(L)1 therapy given for at least 2 doses and either sequential or combined platinum-containing chemotherapy. Patients were required to have an ECOG performance status of 0 or 1, provide a tumor sample for PD-L1 evaluation, and have no previous exposure to, or indication for, EGFR-, ALK-, or ROS1-directed therapy.

Patients were randomly assigned in a 4:4:1 ratio to 1 of 3 treatment arms. In the first group, patients received 20 mg of oral lenvatinib per day alongside 200 mg of pembrolizumab administered intravenously (IV) every 3 weeks for a maximum of 35 cycles. Patients in arms 2 and 3 were treated with 75 mg/m2 of IV docetaxel every 3 weeks, or a daily 24 mg dose of oral lenvatinib, respectively.

Key stratification factors included anti–PD-(L)1 therapy as immediate prior therapy, PD-L1 tumor proportion score (< 50% vs ≥ 50%), and ECOG performance status.

All randomized participants in the intention-to-treat (ITT) cohort were included in the efficacy analysis, while the safety analysis consisted of all randomized participants who received at least 1 dose of study treatment. The graphical method of Maurer and Bretz was utilized to calculate the family-wise type I error rate, which was strongly controlled at a one-sided alpha level of 0.025 across all hypotheses.

The study’s co-primary end points were PFS per RECIST 1.1 by blinded independent central review (BICR) and OS. Secondary end points included overall response rate (ORR) and duration of response (DOR) per RECIST 1.1 by BICR, health-related quality of life, and safety.

Two interim analyses were specified in the study protocol. The first was conducted after the completion of 6 months of follow-up for 200 participants, with a data cutoff of July 14, 2021. This contained a final analysis of ORR for the combination vs docetaxel. At the September 26, 2022, data cutoff of the second interim analysis contained the final analysis of PFS for the combination vs docetaxel and of ORR for the combination vs lenvatinib monotherapy. This occurred after 279 PFS events were observed, and 6 or more months after the last participant was randomly assigned to treatment. Protocol-specified final analysis was conducted after approximately 299 deaths had occurred and 12 or more months had elapsed after the last participant was randomly assigned to treatment. This included final OS data for lenvatinib and pembrolizumab vs docetaxel. The data cutoff for the final analysis was August 11, 2023.

A total of 817 patients were initially screened for the study, 422 of whom were included in the ITT population. Of these 185, 189, and 48 patients were assigned to receive the combination, docetaxel alone, or lenvatinib alone, respectively. Within these respective arms, 181, 177, and 47 received 1 or more doses of the study treatment.

At a median follow-up of 31.8 months (range, 16.5-48.7), 6 patients remain on the combination regimen, 1 patient has completed the regimen, and 174 patients discontinued treatment. Three patients remain on docetaxel, while 174 patients discontinued the drug. One patient remains on lenvatinib alone and 46 discontinued treatment. Reasons for treatment discontinuation across all groups include adverse effects (AEs; 44 with the combination; 42 with docetaxel; 13 with lenvatinib), physician’s decision (2; 13; 0), progression (126; 104; 26) or withdrawal of consent (2; 15; 7).

The median age of patients was 65 years (range, 41-81) in the combination arm, 65 years (range, 30-81 in the docetaxel arm), and 64 years (range, 47-77) in the lenvatinib arm. The majority of patients across all arms were male (63.8% in the combination arm; 68.3% in the docetaxel arm; 60.4% in the lenvatinib arm) and White (78.9%; 71.4%; 72.9%). Most patients also had nonsquamous histology (73.0%; 69.3%; 87.5%), an ECOG performance score of 1 (63.8%; 64.6%; 64.6%), were current or former smokers (83.8%; 87.3%; 85.4%), had a PD-L1 TPS score greater than 50% (73.0%; 74.1%; 70.8%).

A total of 78.4%; 79.4%; and 79.2% of patients in the combination, docetaxel and lenvatinib arms, respectively, received an anti–PD-(L)1 therapy as their immediate prior therapy. Most patients had received prior anti–PD-(L)1 and chemotherapy concomitantly rather than sequentially, and over half of patients in all groups had received pembrolizumab as their only prior PD-L1 inhibitor. Brain metastases were observed in 18.9%, 17.5%, and 20.8% of patients in the combination, docetaxel and lenvatinib arms; the percentage of liver metastases in each respective arm was 22.7%, 20.6%, and 14.6%.

Further efficacy analysis revealed that the lenvatinib combination produced an ORR of 22.7% (95% CI, 16.9%-29.4%), including a 22.2% partial response (PR) rate. ORRs in the docetaxel and lenvatinib arms were 14.3% (95% CI, 9.6%-20.1%) and 12.5% (95% CI, 4.7%-25.2%). This translated to a treatment difference of 8.4% 95% CI, 0.5%-16.3%) between the combination and docetaxel groups and 10.3% (95% CI, -3.1%-19.9%) between the combination and lenvatinib groups. The DOR was 6.9 months (range, 2.3-42.8+) with the combination and 6.8 months (range, 2.5+ to 33.0+) with docetaxel.

The duration of therapy with the combination was 6.1 months (range, 0.03-46.9), followed by 4.4 months with lenvatinib (range, 0.03- 38.5) and 3.4 months with docetaxel (range, 0.03-40.3). Patients underwent a median of 8 (range, 1-35) treatment cycles with the combination and 5 (range, 1-59). with docetaxel. The median internal dose of lenvatinib in the combination was 16.2 mg/day (range, 5-20) vs 19.5 mg/day (range, 10-24) with lenvatinib alone. A total of 80.8% (range, 24%-100%) and 81.4% (range, 43%-100%) of patients in the combination and lenvatinib monotherapy groups received this dosage as a percentage of the planned starting dose.

Additionally, a total of 40%, 52.4%, and 31.2% of patients in the combination, docetaxel, and lenvatinib arms went on to receive subsequent systemic anticancer therapy.

Regarding safety, the combination’s toxicity profile was as expected based on prior studies of the combination in other tumor types. Any-grade treatment-related AEs (TRAEs) occurred in 91.7%, 91.0%, and 89.4% of patients in the combination, docetaxel and lenvatinib groups, respectively. Of these, 59.7%, 48.6%, and 57.4%, respectively, were grade 3 to 5. Serious Aes were seen in 29.8%, 15.3%, and 25.5% of patients, respectively. The percentage of AEs leading to death in the combination group was 3.9%, followed by 1.1% with docetaxel and 6.4% with lenvatinib. AEs leading to treatment discontinuation occurred in 27.6% of patients in the combination group, followed by 14.7% in the docetaxel arm and 12.8% in the lenvatinib arm.

Common TRAEs of any-grade with an incidence of 15% or greater included hypertension (combination = 35.9%; docetaxel = 0.0%; lenvatinib = 36.2%), hypothyroidism, (32.6%; 0.0%; 27.7%), diarrhea (31.5%; 16.4%; 34.0%), asthenia (20.4%; 18.6%; 19.1%), decreased appetite (19.9%; 13.6%; 14.9%), proteinuria (17.1%; 0.0%; 12.8%), stomatitis (17.1%; 10.7%; 21.3%), fatigue (14.9%; 18.1%; 23.4%), nausea (14.4%; 15.8%; 23.4%), decreased neutrophil count (3.3%; 19.8%; 6.4%), anemia (2.2%; 24.9%; 0.0%), and alopecia (1.1%; 25.4%; 0.0%).

“Clearly more effective therapies are needed in this population. It is our hope that in the new year, we will see more studies with innovative agents [that] finally [produce] some positive results so that we can move this area forward,” Leighl concluded.

Dr Leighl disclosed the following relationships: research support paid to the institution from Amgen, AstraZeneca, Bayer, BMS, Eli Lilly, EMD Serono, Guardant Health, Inivata, Janssen, MSD, Novartis, Pfizer, Roche, and Takeda; uncompensated advisory board member for Amgen, AstraZeneca, BioNtech, GSK, Janssen, Merck & Co., Inc., Novartis, and Regeneron; honoraria for CME lectures from Beigene, BMS, Janssen, MSD, Novartis, and Takeda; travel support for CME lectures from AstraZeneca, Eisai, Janssen, MSD, and Sanofi.

Reference

Leighl N, Paz-Ares L, Rodrigues Abreu D, S, et al. Phase 3 LEAP-008 study of lenvatinib plus pembrolizumab versus docetaxel for metastatic non-small cell lung cancer (NSCLC) that progressed on a PD-(L)1 inhibitor and platinum-containing chemotherapy. Ann Oncol. 2023;20(suppl 1):100535. doi: 10.1016/iotech/iotech100535