Smoking History Predicts Efficacy of Immune Checkpoint Inhibitors in Metastatic Non–Small-Cell Lung Cancer

Tumor mutation burden (TMB) has emerged as a potential predictive biomarker for response to immune checkpoint inhibitor (ICI) therapy in patients with non–small-cell lung cancer (NSCLC). However, TMB measurement requires tumor tissue, high expense, and prolonged turnaround time. Results may not be available for early clinical decision-making.

Subscribe to the latest updates from Pulmonary & Critical Care Advances in Motion

Thank you for subscribing!

Error: Please enter a valid email address.

Email Address Submit

Several studies have shown tobacco smoking to be associated with higher TMB in patients with lung cancer.

David C. Christiani, MD, MPH, director of the Christiani Lab in the Division of Pulmonary and Critical Care Medicine at Massachusetts General Hospital, and colleagues recently determined smoking history provides predictive information about the efficacy of ICI monotherapy in patients with advanced NSCLC. In the Journal of the National Cancer Institute, they emphasize that quantifying smoking pack-years provided information additional to a biomarker that's already routinely assessed, programmed death ligand-1 tumor proportion score (PD-L1 TPS).

Methods

The researchers reviewed data on 644 patients with metastatic NSCLC whose tumors underwent successful OncoPanel sequencing and received ICI monotherapy between April 2013 and September 2020. They were divided into:

• 105 never smokers (<100 cigarettes in a lifetime)
• 375 former smokers (quit >12 months before diagnosis)—median smoking pack-years, 28
• 164 current smokers (quit <12 months before diagnosis or still smoking)—median pack-years, 40

OncoPanel reports included TMB and PD-L1 TPS as available.

TMB by Smoking History

There was a dose-dependent relationship between smoking and median TMB:

• Current smokers—12.2 mut/Mb
• Former smokers—9.9 mut/Mb (P<0.001 vs. current smokers)
• Never-smokers—7.6 mut/Mb (P<0.001 vs. former and current smokers)

Outcomes by Smoking History

In multivariable analyses, smoking status was significantly associated with response and progression-free survival:

Overall response rate (ORR)

• Current vs. never smokers—OR, 3.04 (P=0.003)
• Former vs. never smokers—OR, 2.07 (P=0.04)

Progression-free survival (PFS)

• Never vs. former smokers—2.07 vs. 3.65 months; HR, 0.74 (P=0.01)
• Never vs. current smokers—2.07 vs. 3.68 months; HR, 0.60 (P<0.001)

Overall survival (OS)

• Never vs. former smokers—9.9 vs. 12.9 months; HR, 0.85 (P=0.23)
• Never vs. current smokers—9.9 vs. 13.2 months; HR, 0.78 (P=0.10)

Doubling of smoking pack-years was significantly associated with improvement in all three outcomes after adjustment for PD-L1 TPS and other clinicopathological characteristics:

• ORR—OR, 1.21 (P<0.001)
• PFS—HR, 0.92 (P<0.001)
• OS—HR, 0.94 (P=0.01)

In subgroup analysis of patients who had PD-L1 TPS available, the effect of smoking on clinical outcomes was significant only in patients with PD-L1 TPS ≥50%.

Validation of Clinical Utility

Adding pack-years to a baseline prediction model (age, gender, histology, performance status and lines of treatment) improved the model's performance:

• ORR—AUC, 0.67 for the new model vs. 0.64 for the baseline model
• PFS—AUC, 0.65 vs. 0.62
• OS—AUC, 0.73 vs. 0.72

Adding both pack-years and PD-L1 TPS to the baseline model yielded similar results as adding pack-years, PD-L1 TPS and TMB. This suggests pack-years can serve as a surrogate for TMB.

Applying the Findings to Practice

Taking a careful smoking history may aid decision-making about ICI treatment for patients with NSCLC. No practice guidelines exist yet for patients with PD-L1 TPS <50%, and smoking information may be particularly helpful in those cases.

view original journal article Subscription may be required

Explore research in the Division of Pulmonary and Critical Care Medicine

Refer a patient to the Division of Pulmonary and Critical Care Medicine