RE: Smoking History as a Potential Predictor of Immune Checkpoint Inhibitor Efficacy in Metastatic Non-Small Cell Lung Cancer

We read with interest the article in this issue of the Journal by Wang et al. (1), whose study evaluated the association between smoking exposure and treatment response, progression-free survival, and overall survival in patients with advanced non-small cell lung cancer treated with immune checkpoint inhibitor (ICI) monotherapy. The authors provided quantitative evidence that smoking pack-years of tobacco was independently prognostic above clinicopathological characteristics, tumor mutation burden, and PD-L1 expression for response and survival outcomes within a cohort of 644 patients. In this correspondence, we highlight 2 aspects that require further comment: 1) the authors statements on the potential for smoking exposure to inform ICI efficacy, in the absence of a control arm, and 2) the potential importance of smoking pack-years vs smoking status (ie, never, former, or current) to future research.

Single-arm research can only validly distinguish prognostic markers, as opposed to providing conclusions on predictive makers of treatment efficacy, which require randomized control-treatment data (2). Wang et al. (1) report strong evidence of an association between smoking pack-years and prognosis following ICI monotherapy; however, their statements to smoking pack-years being a consistent and readily available marker of ICI efficacy are not possible because of the unavailability of a control arm. Without a matched cohort not using ICI therapy (preferably randomly allocated), the impacts on treatment efficacy are unclear. Notably, prior meta-analyses of ICI randomized control trials have been conducted and do not clearly demonstrate that ICI efficacy differs between never-smokers and former or current-smokers (3,4). Nonetheless, the meta-analyses do leave us contemplating whether ICIs result in superior progression-free survival for all smoker statuses, whether chemo-immunotherapy is essential in never-smokers, and the worth of ICIs in never-smokers with low PD-L1 expression. Further, the meta-analyses fail to provide clarity on never vs former vs current smokers or by pack-years, and thus we appreciate the efforts of Wang and colleagues to provide this information.

Although smoking history has previously been recognized as a prognostic factor for ICIs used in the treatment of advanced non-small cell lung cancer (5,6), Wang et al. (1) provide evidence that smoking packs-years may be a better prognostic marker than categorized smoking history. This is briefly highlighted in Table 2 in Wang et al. (1), which shows that smoker status (ie, never, former, or current) was not statistically associated with overall survival on multivariable analysis; comparatively, Table 3 outlines that smoking pack-years was statistically associated. This provides important insights that reanalysis (preferably meta-analysis) of ICI randomized control trials using smoking pack-years may be valuable to validly answering whether certain individuals are likely to benefit from ICI treatment. We look forward to further insights that can be provided by Wang et al. (1) on the performance of smoking pack-years vs smoking status (eg, Area under the ROC Curves [AUC ROCs] and an optimal cut point of smoking pack-years for clinical practice).

Funding

AR and MJS are supported by Beat Cancer Research Fellowships from Cancer Council South Australia. AMH is supported by a grant from Flinders Foundation (Health Seed Grant).

Notes

Role of the funders: The funders had no role in the design and conduct of the study; interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit.

Disclosures: AR and MJS report investigator-initiated project grants from Pfizer, outside the submitted work. GK, JML, and AMH have no conflicts of interest to disclose.

Author contributions: Writing, original draft—AMH, GK, JML, AR, MJS. Writing, review, and editing—AMH, GK, JML, AR, MJS.

Data Availability

Not applicable.