High tumour mutational burden was associated with more frequent and earlier immune-related adverse events in advanced NSCLC

About one in three patients developed immune-related toxicity

The cohort included patients with advanced NSCLC who received checkpoint inhibitors with or without chemotherapy and whose baseline TMB had been measured with the OncoPanel sequencing assay. Of the 1215 patients, 63.4% received a checkpoint inhibitor alone and 61.4% were treated in the first-line setting.

A total of 421 patients developed an immune-related adverse event. The authors related baseline TMB to these events with two models: logistic regression for whether an event occurred, and Cox proportional hazards regression for how soon it occurred.

A TMB at or above 19 mutations per megabase more than doubled the adjusted odds

Across the cohort, higher baseline TMB carried an adjusted odds ratio of 1.04 for developing an immune-related adverse event (p < 0.001) and an adjusted hazard ratio of 1.02 for earlier onset (p = 0.002).

The authors then compared patients at or above the 90th percentile of TMB, which corresponded to 19 mutations per megabase, with the remaining 90%. The highest-TMB group had an adjusted odds ratio of 2.14 for an immune-related adverse event (p < 0.001) and an adjusted hazard ratio of 1.58 for earlier onset (p = 0.004). The authors report this threshold as the strongest predictor of both outcomes.

Landmark analyses at 6, 9 and 12 weeks addressed treatment duration

TMB is an established predictor of response to checkpoint inhibitors in advanced NSCLC, and patients who respond remain on treatment longer. A raw association between TMB and adverse events could therefore reflect longer exposure alone. The authors report 6-, 9- and 12-week landmark analyses, and conclude that the association is independent of longer immunotherapy exposure.

After adjustment for age, sex, Eastern Cooperative Oncology Group performance status, PD-L1 expression and treatment type, a TMB of 19 mutations per megabase or more remained associated with a higher risk of an immune-related adverse event at each of the three landmarks.

“These findings indicate that a high TMB, particularly TMB more than or equal to the 19 mut/Mb, is associated with an increased risk and earlier irAE onset, independent of longer immunotherapy exposure.”Pecci et al., Journal of Thoracic Oncology (2026)

More CD8+ cells in the highest-TMB tumours

To look for a biological explanation, the investigators measured the density of CD8+, PD-1+, CD8+PD-1+ and FOXP3+ immune cells in tumour tissue with multiplex immunofluorescence. Tumours with TMB of 19 mutations per megabase or more were enriched for CD8+ cells compared with tumours below that threshold (p = 0.04).

The authors interpret this as a sign that the association is biologically driven, and they propose elevated TMB as a biomarker for immune-related adverse events. The abstract does not state whether the other three cell populations differed between the groups.

A threshold drawn from one centre’s own distribution

The 19 mutations per megabase cut-off is the 90th percentile of this cohort, measured on one sequencing panel, OncoPanel, at one institution. The abstract does not say whether the same number identifies the same patients on another assay or in another cohort.

The design is observational. Patients were not assigned to treatment by TMB, and adjusted estimates can correct only for the variables that were measured; for the landmark analyses these were age, sex, performance status, PD-L1 expression and treatment type.