Advanced NSCLC

28P - Generation and validation of a predictive model using pretreatment clinical factors for estimating survival and T790M mutation in EGFR-mutated non-small cell lung cancer in Taiwan

Presentation Number
28P
Speakers
  • C. Lin (Tainan City, Taiwan)
Authors
  • C. Lin (Tainan City, Taiwan)
  • Y. Chou (Tainan City, Taiwan)
  • C. Lin (Tainan City, Taiwan)
  • C. Wu (Taoyuan City, Taiwan)
  • C. Yang (Taoyuan City, Taiwan)
  • J. Chang (Taipei City, Taiwan)
Presentation Topic
Advanced NSCLC

Abstract

Background

Although epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) have been the standard treatment for advanced EGFR-Mutated NSCLC, the generation and validation of a comprehensive platform in predicting survival of these patients remain rare.

Methods

From October 2010 to 2021, we collected potential prognostic factors from advanced stage NSCLC patients receiving EGFR-TKI treatment at National Chen-Kung University, Tainan, Taiwan (NCKUH). Using univariate and multivariate analyses, we identify potential prognostic factors and create a nomogram for risk stratification accordingly. Then we validated the platform in another cohort from Chang Gung Memorial Hospital.

Results

Records of 761 EGFR-Mutated NSCLC patients from NCKUH were retrospectively reviewed. Using univariate analysis, we identified 8 prognostic factors including sex, ECOG status, morphology, mutation, stage, the choice of EGFR-TKIs, and metastasis to liver, brain and multivariate analysis confirmed their independent significance. We established a nomogram based on these factors and successfully classified patients into different risk groups with different survival. This nomogram can be used to predict the possibility of 6-,9-, and 12-month PFS and stratify patients into different risk groups for PFS and OS. In addition, patients with shorter PFS predicted by the nomogram had significantly higher incidence of acquired T790M mutation upon disease progression, which implied the early emergence of T790M might be predicted by this nomogram. We then successful validated the risk score in another cohort including 751 EGFR-Mutated NSCLC patients from Chang-Gung Memorial Hospital. The calibration curves for the probability of survival at 6, 9, and 12 months after EGFR-TKI use revealed a good concordance between the nomogram prediction and actual observation. Moreover, the calibration curves of these two cohorts showed similar pattern.

Conclusions

Our risk stratification can provide additional information to clinicians to evaluate the prognosis and the chance of sequential therapy in patients with EGFR-Mutated NSCLC patients who received targeted therapy.

Legal entity responsible for the study

The authors.

Funding

National Science and Technology Council (110-2314-B-006 -098 -MY3,109-2314-B-006-083 and MOST 108-2314-B-006-092-MY2.

Disclosure

All authors have declared no conflicts of interest.

Collapse