Cisplatin-Free Pembrolizumab Regimen Misses Primary End Point in SMART-KEY

At the 2026 European Society for Medical Oncology (ESMO) Congress, Thiago Bueno de Oliveira, MD, PhD, deputy leader of the Head and Neck Tumor Reference Center and associate physician in the department of Clinical Oncology at A.C. Camargo Cancer Center in São Paulo, Brazil, presented results from the phase 2 SMART-KEY/LACOG 0720 trial (NCT04943445) during a rapid oral session on head and neck cancer.¹ SMART-KEY evaluated a cisplatin-free, pembrolizumab (Keytruda)-based larynx-preservation strategy in patients with locally advanced laryngeal cancer. The 2-year laryngectomy-free survival (LFS) rate was 65.1% (80% CI, 54.9%-73.5%), which was not significantly higher than the 59% historical control rate (P = .46).

Standard larynx-preserving approaches yield similar LFS to one another, without differences in overall survival (OS). However, data have raised concerns about late toxicity and mortality, particularly with concurrent cisplatin and radiotherapy.²

The single-arm, multicenter trial enrolled 43 patients across 12 sites in Brazil. Eligible patients had stage III to IVB squamous cell carcinoma of the larynx and an ECOG performance status of 0 or 1. Patients with T1 tumors and those with T4 tumors penetrating the cartilage or extending more than 1 cm into the base of the tongue were excluded.

Treatment consisted of 3 phases, each given in 21-day cycles to patients without disease progression:

• Induction: 3 cycles of carboplatin (area under the curve 6), paclitaxel 175 mg/m², and pembrolizumab 200 mg.
• Concurrent therapy: 70 Gy of radiotherapy with 3 cycles of pembrolizumab.
• Consolidation: 11 cycles of pembrolizumab.

The primary end point was 2-year LFS. The trial was powered to detect an improvement from 59%, based on the induction arm of the RTOG 91-11 trial, to 75%.

The median age was 61.1 years (range, 42.1-90.6), and 86% (n = 37/43) of patients were male. Most tumors were glottic (63%; n = 27/43) and T3 (84%; n = 36/43). In all, 67% of patients had stage III disease, 23% had stage IVA, and 9% had stage IVB. The PD-L1 combined positive score was 1 or higher in 26% (n = 11/43) of patients.

At a median follow-up of 25.2 months (IQR, 24.4-26.2), 15 patients had an LFS event: 6 underwent laryngectomy and 9 died. LFS did not correlate with PD-L1 expression. At 2 years:

• Larynx preservation rate: 84.3% (80% CI, 68.1%-92.7%).
• Larynx dysfunction–free survival rate: 82.0% (80% CI, 72.4%-88.6%).
• Cumulative incidence of locoregional recurrence: 18.8% (80% CI, 11.8%-27.2%).
• Cumulative incidence of distant recurrence: 7.4% (80% CI, 3.2%-13.9%).
• OS rate: 74.4% (80% CI, 64.7%-81.8%).

Of the 11 deaths, 5 resulted from progression and 6 from adverse events (AEs): 5 from pneumonia or sepsis and 1 from pneumonitis.

In an exploratory comparison with RTOG 91-11, 65.1% of SMART-KEY patients were alive with an intact larynx at 2 years. The corresponding rates in RTOG 91-11 were 58.6% with induction chemotherapy followed by radiotherapy, 66.0% with concurrent cisplatin and radiotherapy, and 52.6% with radiotherapy alone.

Treatment-related AEs occurred in 93% (n = 40/43) of patients and were grade 3 or higher in 42% (n = 18/43). Immune-mediated AEs occurred in 35% of patients, and were grade 3 or higher in 16%. The most common treatment-related AEs were alopecia (46%), odynophagia (35%), hypothyroidism (33%), anemia (30%), and dry mouth, dysphagia, lymphopenia, and weight loss (28% each). Six patients had grade 5 AEs, 1 of which, pneumonitis, was treatment related.

The investigators concluded that SMART-KEY produced 2-year oncologic and functional outcomes numerically higher than the RTOG 91-11 induction arm and comparable with its concurrent cisplatin arm, with low rates of distant recurrence. They added that longer follow-up will show whether the strategy can reduce late morbidity and mortality related to cisplatin and radiotherapy.

Disclosures

Oliveira reported honoraria as a speaker, and consultancy and advisory roles, with MSD, AstraZeneca, Johnson & Johnson, Adium, and Novartis; support for attending meetings or travel from Johnson & Johnson, Adium, and MSD; and institutional research funding from MSD and the Adenoid Cystic Foundation. The study was sponsored by the Latin American Cooperative Oncology Group (LACOG) and funded by MSD.

References

1. William WN Jr, Ferreira da Silva FA, Ramos FBP, et al. A single-arm, multi-institutional, phase 2 study of a pembrolizumab-based organ preservation strategy for locally advanced larynx cancers: SMART-KEY (LACOG 0720). Presented at: 2026 European Society for Medical Oncology Congress; October 23-27, 2026; Madrid, Spain. Abstract 1537RO.
2. Forastiere AA, Zhang Q, Weber RS, et al. Long-term results of RTOG 91-11: a comparison of three nonsurgical treatment strategies to preserve the larynx in patients with locally advanced larynx cancer. J Clin Oncol. 2013;31(7):845-852. doi:10.1200/JCO.2012.43.6097