Prognostic nomogram to predict survival after surgery for synchronous multiple lung cancers in multiple lobes

Academic Article

Abstract

  • © 2014 by the International Association for the Study of Lung Cancer. INTRODUCTION:: In the absence of metastatic disease, surgery for synchronous non-small-cell lung cancers involving multiple lobes can be curative. However, there currently exists no reliable prognostic instrument for this patient population after surgery. We undertook an analysis to examine the prognostic significance of adenocarcinoma histology and developed a prognostic nomogram. METHODS:: This study was a pooled analysis of six previously reported datasets. Patients without extra-thoracic metastasis who underwent surgical resection of synchronous lung cancers in multiple lobes were included. Those with small cell cancer, carcinoid tumor, or exclusively carcinoma in situ were excluded. A multivariable Cox proportional hazards regression model was fitted to identify independent survival predictors for nomogram development. RESULTS:: Data from 467 patients were analyzed. Adenocarcinoma was a sole histology in 253 patients (54.2%). Those with exclusively adenocarcinoma histology had a better median survival than their counterparts: 67.4 versus 36.2 months, (p < 0.001). Multivariable analysis incorporating histology, sex, age, maximal T-size, highest N-stage, and laterality demonstrated that having exclusively adenocarcinoma histology independently predicted an improved survival: hazard ratio 0.61 (95% confidence interval: 0.48, 0.78). Other favorable survival predictors were N0, T-size less than or equal to 3 cm, bilateral cancers, age less than 70 years, and women sex. The developed nomogram was well calibrated and demonstrated a moderate to good discrimination with a bootstrap-corrected Harrell C-statistic of 0.70. CONCLUSION:: Several unique features among patients with resected synchronous multiple lung cancers, including the presence of exclusively adenocarcinoma histology, are of prognostic significance. A simple nomogram incorporating these factors can be utilized to predict patient survival with acceptable accuracy.
  • Authors

    Published In

    Digital Object Identifier (doi)

    Author List

  • Tanvetyanon T; Finley DJ; Fabian T; Riquet M; Voltolini L; Kocaturk C; Bryant A; Robinson L
  • Start Page

  • 338
  • End Page

  • 345
  • Volume

  • 10
  • Issue

  • 2