A predictive model for identifying candidates for adjuvant chemotherapy based on recurrence risk profile of resected,
Timur A Krivitsky1,2, Gavin M Wright3,4,5, Muteb Al Zaidi3,6
1Department of General Medicine, St Vincent's Hospital Melbourne, Fitzroy, VIC, Australia.
Background:
The decision for administering adjuvant chemotherapy (AC) in completely resected node-negative non-small cell lung cancer (NSCLC) is guided by likelihood of disease recurrence or death based on tumor, node, metastasis (TNM) stage. However, within each TNM stage are sub-groups of patients that are more or less likely to relapse than stage alone predicts.
Methods:
In this retrospective cohort study, prospective data from 394 consecutive patients who underwent complete resection of node-negative NSCLC without adjuvant therapies, between 2002 and 2019 was retrospectively analyzed. Independent tumor and host risk factors for recurrence were subjected to multivariate analysis to develop a predictive risk model distributing patients into low-risk or high-risk categories.
Results:
Recurrence risk was independently predicted by a neutrophil:lymphocyte ratio (NLR) of ≥3.5 [hazard ratio (HR), 1.9; 95% confidence interval (CI), 1.1-3.5], visceral pleural invasion (HR, 2.2; 95% CI, 1.3-3.8), histopathology other than adenocarcinoma or squamous cell (HR, 2.6; 95% CI, 1.2-5.5) and tumor size >33 mm (HR, 3.9; 95% CI, 2.3-6.7). The specific combination of risk factors contributed to a score for a risk model which classified 9% of Stage I and 69% of Stage ≥II patients as high-risk. The predicted 5-year disease-free survival (DFS) for high-risk and low-risk patients as scored by the predictive model was 30% and 85%, respectively.
Conclusions:
Our readily reproducible, low-technology model, developed from individually validated tumor/host risk factors, identified sub-groups of resected node-negative NSCLC patients at significantly discordant risk of recurrence to their TNM stage category.
