ERJ Open Research (Dec 2019)

Tumour growth rate as a tool for response evaluation during PD-1 treatment for non-small cell lung cancer: a retrospective analysis

  • Deirdre M.H.J. ten Berge,
  • Daniel P. Hurkmans,
  • Ilse den Besten,
  • Jeroen S. Kloover,
  • Ron H.J. Mathijssen,
  • Reno Debets,
  • Egbert F. Smit,
  • Joachim G.J.V. Aerts

Journal volume & issue
Vol. 5, no. 4


Read online

Background Immune checkpoint inhibitors have emerged as a standard of care treatment for non-small cell lung cancer (NSCLC). To get insight into variations in tumour growth kinetics and their potential predictive values for outcome, we evaluated tumour growth rate (TGR) in patients receiving programmed cell death 1 (PD-1) checkpoint inhibitors. Patients and methods Differences in TGR before and after the start of treatment were calculated by entering the sum of the longest diameters from computer tomography scans before and after the initiation of therapy into a formula that assumes volumetric exponential tumour growth. TGR variations, possible predictors for TGR changes and its relationship to overall survival (OS) were studied. For comparison, tumour response was assessed using Response Evaluation Criteria in Solid Tumours (RECIST) version 1.1. Results Among the 58 evaluable patients, 37 patients (64%) showed deceleration of TGR and 16 patients (27%) showed an acceleration of TGR after initiation of therapy, with a significant difference in median OS of 18.0 months versus 6.0 months (hazard ratio 0.35, 95% CI 0.18–0.71) between these groups. Four patients (7%) were defined as having hyperprogressive disease. In five patients (9%), tumour growth remained stable. These TGR categories were not significantly different according to age, sex, histology, smoking or previous radiotherapy. Of the patients defined as having progressive disease by RECIST version 1.1 at first follow-up, 40% showed response to checkpoint inhibitors by a decrease in TGR. Conclusion Tumour growth kinetics can be used as a clinically relevant predictor for OS in anti-PD-1-treated patients with NSCLC, and may provide additional information to RECIST measurements.