Influence of the number of tumor size measurements on model-derived tumor size metrics and prediction of survival
Sreenath M. Krishnan (1), Lena E. Friberg (1)
(1) Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
Objectives: The tumor size ratio (TSR), time-to-tumor growth (TTG) and tumor growth rate (KG) are frequently suggested predictors of overall survival (OS) for different types of tumors[1,2]. It should however be acknowledged that all available measurements are typically used to estimate these metrics for an individual patient. This study aims to investigate how the number of available tumor size measurements may influence the accuracy of predicting the true tumor size metrics for an individual patient, which in turn could influence the metrics’ value in predicting the hazard of death.
Methods: Tumor size data for 1000 subjects were simulated using a simplified tumor growth inhibition model for bevacizumab+chemotherapy in colorectal cancer[3], at baseline, and at 6,12,18,24,36,48,60,72,84 and 96 weeks. The ‘true’ TSR at week 6 (TSRw6), TTG and KG were derived from the simulated individual profiles and the prospective evaluation function in PsN[4] was applied to investigate the accuracy of the predicted metrics and the OS estimation.
Results: As expected, the accuracy of the tumor size metrics improved as the number of measurements increased. When only baseline and w6 measurements were used in the predictions of TSRw6, about 70% of individuals had <10% deviation from ‘true’ TSRw6. By adding a w12 measurement, the corresponding percentage was 78%. The accuracy in the individuals’ predictions was little affected with addition of later observations. The accuracy in TTG predictions was in general low; the percentage of individuals with <30% deviation from ‘true’ TTG was increased from 32% to 39% when increasing from 2 to 4 observations, while additional measurements only marginally affected the accuracy. The percentage of individuals with <10% deviation from ‘true’ KG improved from 41% (2 observations) to 77% when all 11 observations were used. As the number of tumor size measurements increased, the OS predictions improved as determined by the model fit and the parameters uncertainty. However, to predict the ‘true’ hazard of death, 6 tumor size measurements were needed.
Conclusions: This simulation study demonstrates that TSRw6 is a more promising metric than TTG or KG for early prediction of treatment outcome for an individual patient, since fewer measurements are needed for adequate estimation of the metric and hence for predicting OS, in line with its lower shrinkage [5]. In addition to baseline and a w6 measurement, a w12 measurement appears beneficial for estimating an individual’s TSRw6.
Acknowledgements: This work was supported by the Swedish Cancer Society.
References:
[1] Bruno et al., Clin Pharmacol Ther. 2014.
[2] Bender et al., Br J Clin Pharmacol 2016.
[3] Claret et al., J Clin Oncol. 2013 Jun 10;31(17):2110-4.
[4] Perl-speaks-NONMEM (PsN). [https://uupharmacometrics.github.io/PsN/index.html]
[5] Ribba et al., Clin Pharmacol Ther. 2014.
