Preoperative Prediction of Incomplete Resection in Non-Small Cell Lung Cancer: An Externally Validated Clinical Nomogram

Background
Patients who are surgically treated for stage I-III non-small cell lung cancer have a worse prognosis after incomplete (R1-R2) resection. Predictive factors for completeness of resection have not satisfactorily been established. Our study aimed to develop, and internally and externally validate a prediction model to estimate the risk of incomplete resection, based on preoperative patient-, tumor-, and treatment-related factors.

Method
From a Dutch national database (NKR) all consecutive NSCLC patients diagnosed from 2011 to 2014 who had surgery without neoadjuvant therapy were selected. Fifteen possible predictors were analyzed. Multivariable logistic regression analysis with stepwise backward elimination was used to create a prediction model. Discriminatory ability and calibration of the model was determined after internal validation. External validation was applied in an American dataset from the NCDB, whereupon the model was adjusted. The prediction model was presented as a nomogram.

Result
In the development set of 7,124 patients an incomplete resection was reached in 496 patients (7.0%). Remaining predictors were gender, histology, cT-stage, cN-stage, extent of surgical resection, time interval from diagnosis to surgery, open versus thoracoscopic procedure, and the interaction between procedure and cN-stage. After internal validation, the corrected c-statistic of the resulting nomogram was 0.73. Application of the nomogram to the external dataset of 85,235 patients with R1-R2 resections in 2,485 patients (2.9%) resulted in a c-statistic of 0.70. Calibration revealed good overall fit of the nomogram in both cohorts.


Conclusion
An internationally validated nomogram is presented providing the ability to predict the individual risk of an incomplete resection in patients with stage I-III NSCLC planned for surgery. In case of a relevant probability of incomplete resection, alternative treatment strategies could be considered, such as a larger extent of surgery, neoadjuvant or definitive chemoradiotherapy. In contrast, with a small predicted probability of incomplete resection, the use of surgery is further supported.